{"data":{"slug":"synthexforge-agent-trust-verification-api","name":"Agent Trust, Verification & Machine-Payable Data Utilities API","description":"199 machine-payable x402 endpoints (USDC on Base, chain 8453) across 75 families, plus 140 free siblings. Families: a2a (6), agent (9), agent reputation (2), api (1), audit (8), autonomy (4), bestex (1), bio (13), budget (1), bureau (3), commerce (4), compliance (1), containment (1), contract (1), coordination (1), corpus (1), counterparty (4), credentials (1), data (10), decision (1), delivery (1), dispute (1), edge (2), evidence (2), finance (8), forensics (1), fund (1), governance (2), identity (2), index (6), intel (3), invoice ocr (1), latency (1), licence (1), mandate (1), market (5), mcp (8), memory (2), nav (1), operations (3), output (1), pass (2), payment (2), policy (1), portfolio (1), preflight (1), pretrade (4), procurement (1), provenance (1), quant (4), quota (1), rates (1), readiness (1), receipts (1), research (1), route (1), sec extract (1), security (10), settlement (3), signed research (1), sla (1), sme (1), social (1), spend (2), strategy (2), synthesis (1), task (1), uptime (1), utils (6), verdict (5), verify (7), verify catalog (1), veritrack (1), verticals (2), x402 (3). Price range $0.01-$29 USDC per call. No accounts, no API keys: the request pays for itself via x402. Paid routes answer HTTP 402 with an x402 v2 payment-required envelope (atomic USDC, scheme exact); signed artifacts use Ed25519 + SHA-256.","base_url":"https://x402.synthexforge.com","website_url":"https://synthexforge.com","category":"Data","endpoint_count":182,"min_price_usd":0.02,"uptime_24h":100,"source":"submitted","status":"online","verified":false,"verified_until":null,"payment_ready":true,"payment_ready_until":"2026-11-26T01:40:13.306Z","last_success_at":"2026-11-19T01:40:13.306Z","consecutive_failures":0,"check_interval_minutes":15,"last_checked_at":"2026-11-19T01:40:13.306Z","created_at":"2026-10-04T22:13:04.706Z","uptime":{"24h":100,"7d":100,"30d":99.3,"90d":99.6,"coverage":{"24h":{"daysCovered":1,"checkCount":4},"7d":{"daysCovered":7,"checkCount":538},"30d":{"daysCovered":30,"checkCount":2406},"90d":{"daysCovered":47,"checkCount":3847}}},"avg_response_time_ms":130,"total_checks":3850,"networks":["BSE"],"networks_caip2":["eip155:8453"],"asset":"USDC","endpoints":[{"id":"35055bf6-47a2-4ff5-9e3c-52804522ade6","method":"GET","path":"/api/ebook/download","description":"Download the full ebook 'x402 — The Agent Economy' (Facilitator Press / Synthex Forge): 121 pages, 12 chapters, 19 code blocks on building machine-native services agents pay for over x402. Free metadata at GET /api/ebook, free sample at GET /api/ebook/sample.","mime_type":"application/pdf","is_active":true,"first_seen_at":"2026-10-14T10:47:29.673Z","last_seen_at":"2026-11-19T01:40:12.467Z","variable_amount":false,"min_price_usd":29,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"29000000","price_usd":"29.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"137c41b3-abd9-4ea7-905f-5020391d2fa7","method":"POST","path":"/v1/a2a/conduct-score","description":"A deterministic A2A-LANE CONDUCT score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated task dispositions (date, a2a_conduct_task_delivered, a2a_conduct_task_abandoned, a2a_conduct_duplicate_deliveries, a2a_conduct_retry_storm_tasks, a2a_conduct_protocol_violations) and the engine scores how the agent BEHAVED IN THE AGENT-TO-AGENT PROTOCOL under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the DISPOSITION OF THE TASKS the agent was handed -- the share abandoned, the share delivered more than once, the share that needed three or more attempts, and the share that broke the handoff protocol -- which no sibling does. The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules -- a negative count, a duplicate dated observation, an impossible disposition -- so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported -- never estimated). The score is MONOTONIC: a strictly worse observation set can never score better, every component is a directly-adverse rate and every weight is non-negative, and the published rubric ships with the answer so you can verify that rather than trust it. That is what makes it usable as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.928Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"af34cd3f-2239-4ac8-b577-d4c62668baea","method":"POST","path":"/v1/a2a/counterparty-score","description":"A deterministic A2A counterparty score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, notional_usd, adverse) and the engine scores them under a published rubric onto a 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights volume_depth (0.30), dispersion (0.25), adverse_events (0.25) and recency (0.20). The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative notional, a duplicate dated observation — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-increasing in risk and every weight is non-negative. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.368Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"99053678-7478-4d06-985f-670326ccc790","method":"POST","path":"/v1/a2a/default-probability","description":"A deterministic A2A DEFAULT PROBABILITY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, payment_delay_days, adverse_events) of a counterparty or an asset, and the engine scores the probability that the obligation DEFAULTS under a published rubric onto an ABSOLUTE 0-100 percent scale (higher = more likely to default), with the inverted 0-100 safety score, a probability band (negligible / low / moderate / elevated / high), every component and its raw inputs so you can RECOMPUTE the probability yourself, and a derived confidence. This is the GRADED question: a solvency score says whether a counterparty can meet its obligations, this one says how likely it is to miss the next one — which is what a pricing engine puts in a spread and a credit desk puts in a limit. The scale is ABSOLUTE, not peer-relative, so it does not move when the peer set does. The rubric weights arrears (0.35), coverage (0.25), downside (0.25) and payment_slope (0.15). The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative figure, a duplicate dated observation, a figure outside its published domain — so the probability is NOT DEFINED and is withheld), or `indeterminate` (too few observations to score on, so the probability is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: the guarantee is structural — every weight is non-negative and every component is monotone in the direction the published rubric states. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.660Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e4f0aa68-9967-4680-851c-4b2fdb2aa2e8","method":"POST","path":"/v1/a2a/escrow","description":"Turn a deal between two agents into SIGNED, MACHINE-CHECKABLE release terms. Returns terms_valid | terms_incomplete, with each release condition marked checkable true or false and every gap NAMED. THE RULE: a condition that cannot be machine-checked is reported checkable: false, and the terms are `terms_incomplete` UNLESS an arbiter is named — a human arbiter is exactly the mechanism for a condition no machine can check. NO CUSTODY: this is the terms document, not the vault. We never hold, move or release funds and never sign on a caller's behalf. The terms id is DERIVED from the terms, so a retry cannot mint a second id. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.579Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"4e9e9d1a-ab23-47bb-bd4e-619c3de849e2","method":"POST","path":"/v1/a2a/exposure-score","description":"A deterministic A2A EXPOSURE PROFILE score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated exposure observations (date, exposure_notional_usd, exposure_adverse) against ONE A2A counterparty, and the engine scores the SHAPE of the exposure you are carrying under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the CALLER'S OWN EXPOSURE, which no sibling does: the rubric weights a2a_exposure_notional_concentration (0.30), a2a_exposure_tenor_drift (0.25), a2a_exposure_recency_gap (0.25) and a2a_exposure_adverse_incidence (0.20). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules — a negative notional, a non-binary adverse flag, a duplicate dated observation — so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.945Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"5df46654-8724-4ac3-8c6a-babb1f5f9a26","method":"POST","path":"/v1/a2a/settlement","description":"What does agent A owe agent B for a delegated task, and is the evidence sufficient to settle it? Returns settle | withhold | disputed | insufficient_evidence, with the evidence basis item by item and a settlement INSTRUCTION. THE RULE: no acceptance verdict means `insufficient_evidence`, NEVER `settle` — a settlement without the payee's acceptance is a settlement on somebody's say-so. A `rejected` acceptance means `withhold` WITH THE REASON. A payer who disputes an acceptance means `disputed`, which is neither a settle nor a withhold. The instruction is a RECOMMENDATION: we never hold, move or custody funds and never sign on a caller's behalf. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.978Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b04b66d3-62d8-4e80-af2a-048a8ae0d1e7","method":"POST","path":"/v1/agent-reputation","description":"Compute + sign a deterministic reputation composite ('Credit Bureau of the Agentic Economy') for an agent from its signed outcome events.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.014Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"19ab755a-3e58-4ad7-b88b-34632c33cae3","method":"POST","path":"/v1/agent-reputation/trusted-route","description":"Rank counterparties by reputation (premium routing): deterministic signed trusted-route ranking (best first) over the supplied per-route outcome histories (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.062Z","variable_amount":false,"min_price_usd":0.3,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"300000","price_usd":"0.3000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c8d7f695-8f6d-4e25-9765-48c95d028552","method":"POST","path":"/v1/agent/forensics","description":"Reconstruct what actually happened across a payment and its workflow, and label EVERY statement by how well it is supported. The report separates FOUR classes and never mixes them: VERIFIED_FACTS (proven from signed evidence or the chain itself), PROVIDER_CLAIMS (what a third party asserted, unverified), DERIVED_CONCLUSIONS (computed by us, with the inference stated), and UNRESOLVED (what the evidence cannot settle). An EMPTY evidence set is UNRESOLVED, never an accusation: the absence of evidence is not evidence of wrongdoing. A transaction detail that cannot be read is null with the reason named — never invented. INTEGRITY IS NOT TRUTH: a valid receipt proves the record was signed by that key, not that a provider delivered what it promised. Missing and duplicate events are DETECTED by comparing the expected step list against the recorded evidence, and the diff is reported. An INCONSISTENCY is a fact about the record set; its severity is a separate, labelled judgement for a human to review. READ-ONLY: it mutates nothing, holds no key, contacts no provider and settles nothing.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.488Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"6845d933-5310-40aa-bf9d-8d574ced7f2e","method":"POST","path":"/v1/agent/handoff","description":"Hand a task from one agent to another without losing the thread. Returns handoff_clean | handoff_with_gaps | insufficient_input, with a state digest, the transferred counts, and every risk NAMED with why it is a risk. THE RULE: an `open` item with no owner is a gap and it is LISTED — a handoff that silently drops an unowned open item is how work gets lost between agents. Empty `done` AND empty `open` is `insufficient_input`, because there is no thread to hand over — never a clean handoff. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.985Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"318caeee-486f-4b83-87d3-298bade26940","method":"POST","path":"/v1/agent/preflight","description":"Is this agent workflow READY to execute? A FOUR-state verdict — READY | READY_WITH_WARNINGS | BLOCKED | INDETERMINATE — backed by the checks ACTUALLY PERFORMED, each tied to a requirement in the manifest you supply (the steps, the services each step calls, the agent identity, the planned and maximum spend, the required inputs, the policy, the quota, the evidence rules). READY is NEVER returned because an endpoint is online: every required requirement must have been verified. A check that COULD NOT run is reported indeterminate, is named in checks_not_performed, and caps the verdict at INDETERMINATE — it is never silently downgraded to READY and never folded into BLOCKED. checks_not_performed is mandatory and non-empty whenever anything could not be verified. Every check that observed live state is marked point_in_time, and the response states that a passing pre-flight does NOT guarantee future availability, safety or correctness. Composition, not a second policy engine: the W3 policy gate, the budget gate, the quota gate, the counterparty scorer, the freshness engine, the pre-trade risk engine and the envelope validator are called with YOUR OWN supplied observations. Deterministic: the same manifest against the same state yields the same verdict and the same output digest. ADVICE ONLY: it holds no budget, reserves nothing, settles nothing and cannot stop you calling a paid route directly.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.962Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"2a964da8-5aff-47f1-83c2-47b71d3d686d","method":"POST","path":"/v1/agent/run","description":"Attested Agent Execution — pay $1.00 for a BOUNDED, SINGLE-TURN agent run against a cheap cloud model (default deepseek-v4.1-flash). Hard server-side caps: 45s wall clock (60s absolute max), 8,000 prompt + 2,000 completion tokens, 0 tool calls (v1 exposes no tools), 1 job in flight service-wide (second caller gets HTTP 429 agent_run_busy), and a $2.00/UTC-day upstream spend circuit breaker (HTTP 503 agent_run_daily_budget_exceeded when tripped). Every run returns the result text plus an Ed25519-signed receipt naming the model actually used, measured token/wall usage, and stopped_reason (completed | token_budget | time_budget | tool_budget) — an over-budget run returns its partial result and is NEVER reported as 'completed'. Receipts verify FREE at GET /v1/agent/verify?run_id=<uuid>; the public key is GET /v1/agent/public-key; breaker state is GET /v1/agent/budget.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.222Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"bd24c1cd-9080-462b-8bc1-8a3bb6294397","method":"POST","path":"/v1/agent/treasury/authorize","description":"Authorize a spend and ATOMICALLY RESERVE the budget. The policy is evaluated FIRST, so a refused request never touches the budget and never writes a reservation row. Two SIMULTANEOUS authorize calls against the same budget CANNOT both succeed — proven with real cross-process races, not threads: the live total is re-read inside a BEGIN IMMEDIATE transaction, so the loser sees the winner's row. The refusal carries the REMAINING BUDGET, because a refusal that does not say what is left is not actionable. An Idempotency-Key scoped to (operation, principal) replays the recorded outcome — the same key under a different payer is a DIFFERENT request. Every transition writes a signed, hash-chained evidence record.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.937Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cc4c1693-ca92-4e10-b18e-6664293d27a1","method":"POST","path":"/v1/agent/treasury/budget","description":"Set the budget limit for an agent or customer scope. The limit is a DECIMAL STRING (floats are refused — a float cannot represent a currency amount exactly). Idempotent: setting the same limit twice is the same budget. A limit BELOW the money already committed or reserved is REFUSED, not applied — a budget that is already breached is a finding, not a configuration. THE CAP IS ENFORCED, not advised: the cap arithmetic and the reservation insert share one SQLite transaction, so two concurrent authorizations against the same budget cannot both succeed. TRUST BOUNDARY: if an agent can pay a provider without going through the authorize step, Treasury Guard controls nothing — the platform does not custody the wallet and cannot govern money sent directly to a third party.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.927Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b07495bb-b679-4d71-84d6-b863d001d184","method":"POST","path":"/v1/agent/treasury/commit","description":"Settle a reservation by proving the payment on the PAYMENT RAIL. There is NO success flag in this request: a client-supplied 'the payment succeeded' is NEVER trusted, because the whole product exists to stop a caller asserting a settlement it cannot prove. The rail is read — the settlement ledger and the on-chain receipt oracle — and its verdict decides the outcome. A TIMEOUT IS NOT A FAILURE: an unreadable rail produces AMBIGUOUS, and a commit on an AMBIGUOUS transaction is REFUSED until it is reconciled. A payment reference already used by another transaction is detected as an INCIDENT and the budget is NOT released. TRUST BOUNDARY: if an agent can pay a provider without going through the authorize step, Treasury Guard controls nothing.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.346Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"23041a10-e817-4295-943e-ab5ef9b7ff33","method":"POST","path":"/v1/agent/treasury/release","description":"Release a RESERVED reservation and return its whole amount to the budget — or RECONCILE an AMBIGUOUS transaction against the payment rail (action='reconcile'). The reconciliation reads the rail and resolves to SETTLED, RELEASED, or stays AMBIGUOUS. It NEVER GUESSES: an unreadable chain resolves to nothing, and no retry is authorised. Only a PROVEN non-payment authorises a retry, which is what makes the ambiguous path safe. Every transition writes a signed, hash-chained evidence record.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.991Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3ed40c0c-110e-4859-bd6b-f465e74587e8","method":"GET","path":"/v1/agent/treasury/report","description":"The budget and usage report for one agent or customer scope: the limit, the committed and reserved amounts, the remaining budget, the transactions by state, and the INVARIANT the report is about — 'committed + reserved <= limit'. The report says so explicitly when that invariant does not hold, rather than presenting the numbers as fine. Budgets are ISOLATED by scope: one agent's spend never appears in another's budget. The report also carries the TRUST BOUNDARY and the list of what this product does and does not enforce.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.439Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"831b534e-e594-41e6-aaec-540765f4d803","method":"POST","path":"/v1/audit/conduct-score","description":"A deterministic AUDIT-LANE CONDUCT SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated audit observations of how ONE counterparty or asset actually behaved (date, audit_conduct_settled_ratio_pct, audit_conduct_disclosure_defect_count, audit_conduct_dispute_resolution_days, audit_conduct_attestation_age_days, audit_conduct_observed_notional_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is cleaner conduct) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_conduct_settlement_integrity (0.30), audit_conduct_disclosure_adequacy (0.25), audit_conduct_dispute_containment (0.20) and audit_conduct_attestation_currency (0.25). It is NOT a rename of /v1/mcp/conduct-score (the MCP/A2A interaction record), of /v1/counterparty/default-probability (the fulfilment-performance record), of /v1/counterparty/exposure-score (the exposure the holder carries), of /v1/counterparty/solvency-score (the obligor's own obligation ladder), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative notional, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.609Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"72c3de64-8d10-4b27-b914-88101f1b5e4f","method":"POST","path":"/v1/audit/counterparty-score","description":"A deterministic AUDIT-LANE COUNTERPARTY SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of how ONE counterparty or asset CLOSED OUT what went wrong (date, audit_remediation_open_days, audit_remediation_verification_passed, audit_remediation_recurrence_count, audit_remediation_evidence_item_count, audit_remediation_remediated_notional_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is cleaner closure) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_remediation_closure_latency (0.30), audit_remediation_verification_pass_rate (0.25), audit_remediation_recurrence_rate (0.20) and audit_remediation_oversight_evidence_completeness (0.25). It is NOT a rename of /v1/audit/conduct-score (the audit-quality of the conduct), of /v1/mcp/conduct-score (the MCP/A2A interaction record), of /v1/counterparty/default-probability (the fulfilment-performance record), of /v1/counterparty/exposure-score (the exposure the holder carries), of /v1/counterparty/solvency-score (the obligor's own obligation ladder), of /v1/pretrade/counterparty-score (a pre-trade gate on one position), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative notional, a duplicate dated observation, a verification share outside 0..100 - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.831Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"5ce46cbf-9d5f-4c7b-ad50-0442188b3b06","method":"POST","path":"/v1/audit/default-probability","description":"A deterministic AUDIT-LANE DEFAULT-PROBABILITY EVIDENCE SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated audit-sample observations of how strong the default-probability signal is inside the population your audit actually reviewed (date, audit_default_probability_reviewed_entity_count, audit_default_probability_flagged_entity_count, audit_default_probability_reviewed_notional_usd, audit_default_probability_impaired_notional_usd, audit_default_probability_evidence_lag_days, audit_default_probability_uncorroborated_flag_count) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is a cleaner default-probability evidence base) with a risk band (minimal / moderate / elevated / severe), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_default_probability_default_signal_incidence_ratio (0.30), audit_default_probability_impairment_depth_ratio (0.25), audit_default_probability_evidence_staleness_ratio (0.25) and audit_default_probability_obligor_support_absence_ratio (0.20). It is NOT a rename of /v1/pretrade/default-probability (a pre-trade gate on one position), of /v1/mcp/default-probability (default event counts and a payment delay on an MCP interaction record), of /v1/index/default-probability (an index constituent set), of /v1/counterparty/default-probability (the counterparty's own fulfilment record), of /v1/audit/conduct-score, of /v1/audit/counterparty-score, of /v1/audit/solvency-score, of /v1/audit/issuer-score, of /v1/audit/risk-score, of /v1/audit/exposure-score nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative count, a negative notional, a negative evidence lag, a duplicate dated observation, a flagged count above the reviewed count, an uncorroborated count above the reviewed count, an impaired notional above the reviewed notional - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.907Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"4456b153-38eb-4ffb-ba77-6a64cac31665","method":"POST","path":"/v1/audit/evidence-package","description":"Assemble and sign ONE self-contained evidence package for a third party: a Merkle root over the canonicalised artifact set (order-independent, with per-artifact proofs), a grade for every assertion (SUPPORTED / PARTIAL / UNSUPPORTED / NOT-ASSESSABLE) with the basis that decided it, and a named gap for every missing piece of evidence. Missing evidence is NOT-ASSESSABLE plus a gap, never a guess. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.420Z","variable_amount":false,"min_price_usd":4,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"4000000","price_usd":"4.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"fa904c6a-bf48-47ab-a041-8b0a7d8a9570","method":"POST","path":"/v1/audit/exposure-score","description":"A deterministic AUDIT-LANE EXPOSURE SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated audit observations of how much of the AUDITED NOTIONAL was NOT PROTECTED (date, audit_exposure_audited_position_count, audit_exposure_breach_position_count, audit_exposure_audited_notional_usd, audit_exposure_unperfected_notional_usd, audit_exposure_enforcement_avoided_notional_usd, audit_exposure_diluted_notional_usd, audit_exposure_restricted_notional_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is better protected) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_exposure_perfection_deficit_ratio (0.30), audit_exposure_enforcement_avoidance_ratio (0.25), audit_exposure_post_audit_dilution_ratio (0.25) and audit_exposure_transfer_restriction_ratio (0.20). It is NOT a rename of /v1/audit/conduct-score (the audit-quality of the conduct), of /v1/audit/counterparty-score (the remediation and closure track record), of /v1/audit/solvency-score (the solvency coverage of the audited sample), of /v1/audit/issuer-score (what the auditor concluded about an issuer's reporting), of /v1/audit/risk-score (the control and limit environment), of /v1/pretrade/exposure-score, of /v1/index/exposure-score, of /v1/counterparty/exposure-score, of /v1/a2a/exposure-score nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative count, a negative notional, a duplicate dated observation, a breach-position count above the audited-position count, a subset notional above the audited notional - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.760Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e5f49977-3669-4737-b291-d4d942c60513","method":"POST","path":"/v1/audit/issuer-score","description":"A deterministic AUDIT-LANE ISSUER SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of what an AUDITOR CONCLUDED about an ISSUER's REPORTED POSITION (date, audit_issuer_audited_period_count, audit_issuer_modified_opinion_period_count, audit_issuer_scope_limitation_period_count, audit_issuer_restated_period_count, audit_issuer_evidence_age_days) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is better-reported) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_issuer_opinion_modification_ratio (0.30), audit_issuer_scope_limitation_ratio (0.25), audit_issuer_restatement_pressure (0.25) and audit_issuer_evidence_age_days (0.20). It is NOT a rename of /v1/audit/conduct-score (the audit-quality of the conduct), of /v1/audit/counterparty-score (the remediation and closure track record), of /v1/audit/solvency-score (the solvency coverage of the audited sample), of /v1/mcp/issuer-score (the issuer's paper standing), of /v1/index/issuer-score (an issuer set as a set), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative period count, a duplicate dated observation, an adverse count above the audited-period count - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.971Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"22edaea4-e34c-4a95-87ec-de2ac5a3f4d4","method":"POST","path":"/v1/audit/risk-score","description":"A deterministic AUDIT-LANE RISK SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of the CONTROL AND LIMIT ENVIRONMENT behind a counterparty's or an asset's operations (date, audit_risk_control_tested_count, audit_risk_control_deficient_count, audit_risk_observed_notional_usd, audit_risk_limit_breach_notional_usd, audit_risk_period_count, audit_risk_model_drift_period_count, audit_risk_policy_exception_period_count) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is better-controlled) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_risk_control_deficiency_ratio (0.30), audit_risk_limit_breach_notional_ratio (0.25), audit_risk_model_drift_ratio (0.25) and audit_risk_policy_exception_ratio (0.20). It is NOT a rename of /v1/audit/conduct-score (the audit-quality of the conduct), of /v1/audit/counterparty-score (the remediation and closure track record), of /v1/audit/solvency-score (the solvency coverage of the audited sample), of /v1/audit/issuer-score (what the auditor concluded about an issuer's reporting), of /v1/mcp/issuer-score (the issuer's paper standing), of /v1/index/issuer-score (an issuer set as a set), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative count, a negative notional, a duplicate dated observation, a deficient-control count above the tested-control count, a breach notional above the observed notional, a drifted-model or policy-exception period count above the audited-period count - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.423Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"75fc9df2-67c5-4578-af97-2d2363d04a01","method":"POST","path":"/v1/audit/solvency-score","description":"A deterministic AUDIT-LANE SOLVENCY SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of how much of an AUDITED POPULATION's SOLVENCY COVERAGE was actually PROVEN (date, audit_solvency_sampled_entity_count, audit_solvency_covered_entity_count, audit_solvency_evidence_age_days, audit_solvency_reviewed_notional_usd, audit_solvency_uncovered_notional_usd, audit_solvency_obligor_datum_present_pct) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is better-proven coverage) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights audit_solvency_sample_coverage_deficit (0.35), audit_solvency_uncovered_notional_ratio (0.25), audit_solvency_evidence_staleness (0.20) and audit_solvency_obligor_datum_completeness (0.20). It is NOT a rename of /v1/audit/conduct-score (the audit-quality of the conduct), of /v1/audit/counterparty-score (the remediation and closure track record), of /v1/mcp/solvency-score (the balance-sheet ratios), of /v1/counterparty/solvency-score (the obligor's own obligation ladder), of /v1/index/solvency-score (an index constituent set), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative notional, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.265Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"9f3b2db2-54e0-4310-a46f-32e7cd5f16ae","method":"POST","path":"/v1/autonomy/commitment-check","description":"Did this agent keep the commitment it made? Reads the supplied commitments and the supplied event log, and grades each commitment met / not_met / unclear / unverifiable with a reason that quotes what was read. THE RULE: only an EXPLICIT LINK between a commitment and an observation can carry a `met` or a `not_met` — a match the ENGINE inferred from text can only carry `unclear`, because a guess must never produce a pass. This judges an agent's promise about its OWN conduct: it is NOT a payment deadline on a rail and NOT a vendor SLA. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.811Z","variable_amount":false,"min_price_usd":0.15,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"150000","price_usd":"0.1500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"65e7c7dc-a950-4ba5-8cd8-cf6abbf9aa93","method":"POST","path":"/v1/autonomy/deadlock-detect","description":"Are two or more agents waiting on each other, and what breaks the cycle? Reads the supplied WAIT-FOR GRAPH — edges of the form 'agent A waits on agent B' — finds every cycle with an explicit DFS, and names the cheapest edge that breaks each one (ranked optional, external, internal, required). THE RULE: an EMPTY edge list is `unverifiable`, never `no_deadlock` — 'no cycle was found in an empty graph' and 'the graph is acyclic' are different claims, and only the first is true. A cycle of ALL required waits is reported with the cost of breaking it. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.820Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"7f5b74a0-ca27-46ff-b302-a4a5c10bd54d","method":"POST","path":"/v1/autonomy/delegation-evaluate","description":"Was this task delegation well-formed? Grades four dimensions of the supplied delegation — scope, authority, verifiability, blast radius — as met / not_met / unclear / unverifiable, with the signal matched for each. THE RULE: an UNSTATED authority is never permissive, and an UNSTATED blast radius is never invented — this engine does not supply a number the caller did not. A delegation whose dimensions could not all be read is `incomplete`, NEVER `well_formed`. This asks whether the DELEGATION INSTRUMENT is sound, which is a different question from whether the work in progress can be handed over. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.773Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b4344ec8-240a-4cda-b387-5def8df62ee1","method":"POST","path":"/v1/autonomy/workflow-recovery","description":"Where does this failed workflow resume? Reconstructs the last-known-good state from an ordered step log and emits completed / incomplete / duplicated / missing steps plus a safe resume point and the instructions to resume from it. DUPLICATE DETECTION KEYS ON BOTH the idempotency key AND the payload hash: the same key with the same payload is an operation that ran twice, and the same key with a DIFFERENT payload is a key-reuse hazard — reported separately, because a key can only authorise one operation. A step that is `in_progress` is NOT safe to re-run — its side effect may have landed, and re-running it is how a payment happens twice — so it is flagged `check_before_retry`. An unrecognised status is never assumed to be pending. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.131Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"69322ac0-ca97-4581-97ef-a6a867d0546c","method":"POST","path":"/v1/bestex/evidence","description":"A MiFID-style best-execution attestation: proof that an order was executed in line with the mandate. Returns within_tolerance / outside_tolerance / insufficient_evidence, with the implementation shortfall, the slippage, the gap to every benchmark supplied, the timing risk and a quality grade. Implementation shortfall is undefined without an arrival price, so a missing one returns `insufficient_evidence` naming the field — the engine never substitutes a fill or a guess. Fills that do not sum to the notional within 1bp are `insufficient_evidence` naming the gap. Every bps figure is reproducible from the numbers shown: the arithmetic is written into the findings. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.963Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8383df88-aace-47ff-a8f3-95130fabcc21","method":"POST","path":"/v1/bio/audit/receipts","description":"Price $0.25 / 250000 atomic per receipt. A signed, tamper-evident RECORD that a named administrative biosecurity check was carried out on a named subject, with the hash of the inputs so a third party can prove the record was not edited. The subject is HASHED, never stored. The outcome_category is the CALLER's own label, reproduced verbatim — this is NOT a biosafety determination, a regulatory decision or a scientific evaluation, and the receipt carries its limitations. Ed25519-signed and SHA-256 reproducible, verifiable offline by anyone. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.753Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"950b0046-e518-4f63-9124-b292d445a5ee","method":"POST","path":"/v1/bio/audit/verify","description":"Price $0.02 / 20000 atomic per verification. Verify a BioAudit receipt a THIRD PARTY issued: the receipt's Ed25519 signature and SHA-256 digest are checked against the public key embedded in the receipt itself. The verdict establishes only that the receipt has not been TAMPERED WITH — it does NOT establish that the underlying check was performed, was correct, or was performed by anyone competent. A receipt that does not verify is a NEGATIVE ANSWER (valid: false with the reason), not an error. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.513Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ad94936e-edf8-48f7-9c73-f4f2e04b2c4d","method":"GET","path":"/v1/bio/catalogue","description":"Price $0.02 / 20000 atomic per catalogue read. The machine-readable index an AGENT routes from: which BioTrust services exist, what each costs, what input schema it takes, what it will not do, and which are callable (a `proposed` service is a plan, not a route). The catalogue is DERIVED from the live route table and the price table, so it cannot name a route that is not on the app. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.660Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"2d76387f-ed5f-4c8c-a0f0-d86d85db2861","method":"POST","path":"/v1/bio/compliance/query","description":"Price $0.50 / 500000 atomic per query. Retrieve CITED passages from an official biosafety-guidance corpus built at deploy time: every answer carries a publisher, a URL, a version, a jurisdiction, a licence, a retrieval date and a freshness verdict. Where two sources disagree the disagreement is REPORTED and NEITHER is selected. `compare` diffs two document versions and reports the differences without interpreting them. This is retrieval of published text, NOT legal advice, and it confers no compliance status. Deterministic, no LLM, no network at request time.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.431Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"62fb0b02-fcf6-4d2a-bc54-e09335f3adc2","method":"POST","path":"/v1/bio/labassets/reconcile","description":"Price $0.50 / 500000 atomic per reconciliation. Reconcile declared custody records against their own declared history across seven checks — quantity variance, missing approval, custody break, overdue audit, duplicate item, unclassified item and location mismatch — and produce an accountability report with a SHA-256 digest over it. `item_class` is a CLOSED vocabulary, and the schema carries NO free-text field; unknown keys are refused rather than ignored. This reconciles RECORDS only: no physical verification, no chain-of-custody legal attestation, and a finding indicates a record defect, NOT missing material. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.211Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"56b99b28-91a6-417d-97de-304b5ae44f04","method":"POST","path":"/v1/bio/labcyber/events","description":"Price $0.50 / 500000 atomic per pass. Run EIGHT DOCUMENTED rules over the caller's own security events — privilege change, after-hours access, geo jump, MFA failure burst, mass resource access, config drift, dormant reactivation and export volume spike — with the evidence event ids, the window and a rationale on every finding, plus a SHA-256 digest over the finding set. The rule set is FIXED: a caller may not inject a rule. A request carrying a credential-bearing key is REFUSED — this service does not ingest credentials. This is rule-based detection over caller-supplied telemetry ONLY: it is not a SIEM, not an EDR, and the absence of a finding is not evidence of security. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.147Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"4af2807d-16cb-4c7d-802d-fa46126c29cd","method":"POST","path":"/v1/bio/outbreak/events","description":"Price $0.05 / 50000 atomic per query. Normalise official public-health notices from two verified publisher feeds (WHO Disease Outbreak News and UKHSA news) into deduplicated, revision-tracked, freshness-flagged events an agent can filter and cite. A publisher's revision updates the SAME event; the same story reported by two publishers is ONE event carrying BOTH source references. Where the two publishers disagree on a date or a category, the disagreement is REPORTED and NEITHER value is selected. A source that cannot be reached is reported as `provider_unavailable` and the response is `degraded` — the working sources are still returned. This is aggregation of PUBLISHED notices: no independent outbreak detection, no case-count verification, and NOT a substitute for official guidance. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.853Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"dc91d771-b6b3-4ff3-88b6-ceb54cb3f398","method":"POST","path":"/v1/bio/provenance/verify","description":"Price $0.05 / 50000 atomic per verification. Prove the origin, integrity and version history of an approved dataset from the manifest the caller supplies: hash well-formedness, name uniqueness, attribution metadata, lineage with no orphans and no cycles, version continuity, and a reproducible root hash. NO FILE IS UPLOADED and no dataset content passes through the service — only hashes and metadata. A valid hash proves the bytes the caller hashed; it does NOT prove the data is scientifically valid, correctly licensed for the caller's use, or free of errors. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.174Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8a6a8c6b-fab8-49b0-ba13-80d8f5e53370","method":"POST","path":"/v1/bio/research-reviews","description":"Price $1.50 / 1500000 atomic per recorded review. A structured dual-use review organised by qualified humans: an evidence register, an assigned reviewer roster, justified decisions, and an escalation path. A category carrying `escalation: true` RAISES the quorum and the change is recorded; the reviewer may not be the creator; a decision needs a justification; and a review cannot close until the required number of DISTINCT reviewers have decided. The category vocabulary is a nine-label CHECKLIST, not a risk-assessment methodology, and it carries NO technical method content. This is NOT a DURC or IBC process, and no decision made here is a safety determination. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.320Z","variable_amount":false,"min_price_usd":1.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1500000","price_usd":"1.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"17c1ec18-7add-4ab8-a33b-c75e0f1c8d48","method":"POST","path":"/v1/bio/risk/assess","description":"Price $2.00 / 2000000 atomic per declared-setup review record. A STRUCTURED REVIEW RECORD over a declared institutional setup: the caller declares the work type, the agent risk group, the containment level, the activity modifiers and the controls already in place, and the service reports which controls the published guidance associates with that declaration, which of them are NOT declared, and what a qualified human must still verify. It does NOT assess biological risk, does NOT score or rank any pathogen or agent, does NOT certify containment, and does NOT replace a biosafety officer, an Institutional Biosafety Committee or an institutional risk assessment. The request schema is ENUM-ONLY: there is no field for a pathogen name, a sequence, a method, a quantity or free text, and an attempt to supply one is refused. An `insufficient_information` result is a CORRECT answer, not an error: it means the declaration was incomplete and the service names what was missing rather than guessing it. Every rule in the control matrix carries a citation that resolves to a document in the published compliance corpus. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.032Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"662b0075-97a2-459d-b6d1-3955d7cf2109","method":"POST","path":"/v1/bio/trust/providers","description":"Price $1.00 / 1000000 atomic per registry record. A registry of provider claims that can be checked: a signed statement, a methodology, an evidence link, and an explicit line between what a provider SAYS and what a signature PROVES. A claim may be labelled `verified` only when its Ed25519 signature verifies over the canonical JSON AND it carries a methodology AND an evidence link; otherwise the registry DOWNGRADES it to `self_reported` and records why. `verified` means 'signature-verified claim by a named key holder' and nothing more — not true, not safe, not audited, not certified. There is NO score, NO ranking and NO star rating. No independent audit is performed. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.112Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"1f5e0f7d-c0e4-4da6-ba63-17c68752eaa4","method":"POST","path":"/v1/bio/verification/checks","description":"Price $0.75 / 750000 atomic per verification check. Verify the EXISTENCE and STANDING of a researcher, an institution, a publication or a company against the public ORCID, ROR, OpenAlex and UK Companies House registries, and screen a name against the official OFAC, EU and UK OFSI consolidated sanctions lists. Every answer carries a citation, a retrieval timestamp and the registry snapshot date. A valid ORCID, ROR id, DOI or company number proves an IDENTIFIER EXISTS — it does NOT prove a person is trustworthy, a supplier is safe, or a paper is sound. The sanctions screen surfaces CANDIDATES for human review: it is NOT a legal determination, NOT a compliance programme, and NOT advice — there is no sanctioned verdict and no safe verdict, by design. Personal identity-document (KYC) verification and professional-credential verification are NOT performed by this service. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.256Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"7e3174b9-252f-41a9-bc49-477860fc7c62","method":"POST","path":"/v1/bio/workflows","description":"Price $0.50 / 500000 atomic per recorded workflow decision. A documented biosafety or biosecurity review, recorded as an ADMINISTRATIVE record: who approved what, when, on which policy version, and with what separation of duties. The state machine REFUSES an illegal transition rather than ignoring it, `approve` requires an actor other than the creator, `close` requires every required role to have approved, and the history is append-only. This records and routes a HUMAN decision process; it does NOT perform or validate the risk assessment, and an approval here is NOT evidence that a facility is safe. Deterministic, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.990Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"221dedcd-a6ea-46ab-8a25-9c8dd8edc785","method":"POST","path":"/v1/budget/preflight","description":"Pay-per-decision agent-safety gate an agent calls BEFORE any spend. Caller supplies {agent_id, action_amount_usd, policy {per_tx_cap, daily_cap, spent_today}}; the engine deterministically evaluates the proposed amount against the policy and returns an Ed25519-signed, SHA-256-reproducible allow/deny with remaining_budget_usd. Safe-deny on malformed input; the gate never settles or moves funds (caller-attested budget state). Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.221Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"7f1d1e1f-45cd-4b65-9775-829f6410928e","method":"POST","path":"/v1/bureau/audit","description":"x402 Verification Bureau — full conformance audit of ANY target endpoint against standard v1.0.0 (every clause C1..C11), dated, Ed25519-signed, coverage gaps named explicitly. SSRF-guarded. Free preflight at POST /v1/bureau/preflight; signature verifiable FREE at POST /v1/bureau/verify.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-14T14:16:42.398Z","last_seen_at":"2026-11-19T01:40:13.138Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"89d59183-24fd-49cf-896a-27b51668700c","method":"POST","path":"/v1/bureau/receipt-check","description":"x402 Verification Bureau — conformance-check ANY issuer's signed receipt/offer (not just ours) against standard v1.0.0. Accepts third-party artifacts: structural conformance always runs; cryptographic verification runs against the declared public key when it is ours and is honestly reported as not_observed when we hold no key for the issuer. Dated, Ed25519-signed result. Verify it FREE at POST /v1/bureau/verify.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-14T14:16:42.398Z","last_seen_at":"2026-11-19T01:40:11.719Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"98981b93-dad5-4475-a346-231d18f2189b","method":"POST","path":"/v1/bureau/reputation","description":"x402 Verification Bureau — reputation score for an endpoint or issuer computed deterministically from caller-supplied SIGNED observation history ({observed_at, verdict|outcome, ...}). Unverifiable events are excluded and NAMED in coverage_gaps — never silently counted. Dated, Ed25519-signed report, verifiable FREE at POST /v1/bureau/verify.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-14T14:16:42.398Z","last_seen_at":"2026-11-19T01:40:12.798Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"106d85f8-c863-4f23-a10c-5a32aff19cd4","method":"POST","path":"/v1/commerce/deliverable-acceptance","description":"Is the delivered artifact set the deliverable? Checks every requirement on four dimensions — PRESENCE, INTEGRITY (digest), SIZE and FORMAT — and returns accept | reject | accept_with_notes with a per-requirement row. THE RULE: an absent expectation is UNDETERMINED, never a pass — a requirement whose expected hash, size or format the caller never supplied is undetermined on that dimension, so the acceptance is qualified (`accept_with_notes`) and the notes name exactly what was NOT verified. A malformed hash is refused rather than compared, because comparing it and calling it a match would certify an integrity check that never happened. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.581Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3baa1815-81ef-460a-8a27-c121fdb768a4","method":"POST","path":"/v1/commerce/dispute-evidence","description":"Assemble the evidence record for a disputed transaction. Reports each section — parties, transaction, timeline, proof, contract — as present / missing / unverifiable / contradicted, and returns the record's strength: strong, partial, insufficient or contradicted. THE RULE: a CONTRADICTION is not a missing item — missing is 'you did not give it to me', contradicted is 'you gave me two things that cannot both be true', and the second is strictly worse. An artefact that carries no digest, signature, reference or content is `unverifiable`, never `present`: a claim of proof that carries no proof is not proof. This is the step UPSTREAM of a dispute ruling — it assembles the record a ruling would consume. It fetched no settlement, receipt or contract from any system. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:10.541Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"eb043aaf-2a60-4155-825c-b2a74c24b85d","method":"POST","path":"/v1/commerce/intent-verify","description":"Does this proposed order match the original instruction? Compares quantity, amount, currency, recipient, category, deadline and description field-by-field and returns field | expected | actual | severity per dimension. THE RULE: an absent field is UNVERIFIED, never agreement — and an unverified amount, currency, recipient, quantity or category means the order is `incomplete`, NOT a match, because an order whose amount was never compared cannot be certified as matching an instruction. Amounts are compared as exact decimals, never floats, and a declared tolerance is applied AND reported. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.882Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"bac2f36d-8fb9-4a09-a0ee-906b9e6136b4","method":"POST","path":"/v1/commerce/warranty-verify","description":"Is this warranty claim valid against the terms the caller supplies? Tests the claim against five requirements — coverage, period, exclusions, obligations, timeliness — returning met / not_met / unclear / unverifiable per requirement with the clause cited. THE RULE: a term that is SILENT leaves the requirement `unclear` or `unverifiable`, never `met` — there is no code path from 'we could not check this' to `valid`. A verdict of `indeterminate` is a first-class outcome: the claim is neither validated nor refused. This service evaluated ONLY the terms and the claim the caller supplied; it consulted no warranty database or register and is not legal advice. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.570Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c626ffb0-5672-4e8c-8be3-e09092dff4fd","method":"POST","path":"/v1/compliance/check","description":"Classify an AI use-case against EU AI Act Annex III risk classes and list applicable obligations.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.845Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"674ce93b-40ed-4e28-a886-6b2ed9aec995","method":"POST","path":"/v1/containment/register","description":"Register an agent/custodian's emergency-contact DID, declared policy, and signed 'if condition X, notify/throttle' rules; Synthex Forge ATTESTES readiness and RELAYS signal directives (Ed25519 + SHA-256). Never holds kill-switch execution authority.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.285Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f121dacf-a2a4-44b7-b013-ca7116130aa5","method":"POST","path":"/v1/contract/term-extract","description":"Extract the clauses that decide money and liability from a commercial agreement, by RULE over the supplied text (no LLM in the answer path): parties, effective/expiry dates, term and auto-renewal, price and escalation, payment terms, liability caps and exclusions, termination rights and cure period, governing law and venue. Every clause that cannot be located deterministically is reported as an AMBIGUITY WITH A REASON — never guessed. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.109Z","variable_amount":false,"min_price_usd":1.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1500000","price_usd":"1.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8f0eda94-b366-4d2f-af5c-f42d54341cd8","method":"POST","path":"/v1/coordination/negotiate","description":"Structured offer / counter-offer rounds. Records each round against the declared constraints — price, currency, deadline, duration and mandatory terms — and emits a MACHINE-READABLE AGREEMENT when an acceptance satisfies them. THE RULE: a round is IDEMPOTENT on round_id. A replayed round returns the original result unchanged and is not counted twice, because a retried round must not move a price twice. A constraint kind that was NOT declared is not checked, and the response names the declared and undeclared kinds so 'nothing was violated' is never read as 'nothing applied'. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.232Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cc80ea16-8d86-465d-9654-2080715ecf3d","method":"POST","path":"/v1/corpus/anchor","description":"Dated, signed proof that a document set existed. A Merkle root over the corpus, the leaves, and a REAL inclusion proof for the first leaf that a verifier can recompute. The leaves are sorted BY NAME, so the root is ORDER-INDEPENDENT and a second party reproduces it without being told our conventions. Every leaf is RECOMPUTED from the supplied bytes where bytes exist, and a declared hash that contradicts its own bytes is `insufficient_input` NAMING the document — no root is computed over a leaf that does not hash to what was declared. On-chain notarisation is NEVER performed: `notarised` is always false. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.191Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c47309c1-2e22-4750-966c-c5a734eff0ac","method":"POST","path":"/v1/counterparty/default-probability","description":"A deterministic COUNTERPARTY-DEFAULT-PROBABILITY-SHAPED score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty's fulfilment-performance record (date, counterparty_default_probability_fulfilment_ratio, counterparty_default_probability_maximum_gap_days, counterparty_default_probability_renegotiation_count, counterparty_default_probability_partial_completion_pct) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the COUNTERPARTY'S OWN dated fulfilment record -- how reliably it completed what it owed, how deep its misses ran, how often it renegotiated instead of paying, and how much of its own cures stuck -- which no sibling does: the rubric weights counterparty_default_probability_deterioration_trend (0.30), counterparty_default_probability_stall_duration (0.25), counterparty_default_probability_fulfilment_shortfall (0.25) and counterparty_default_probability_recovery_failure (0.20). It is NOT a rename of /v1/pretrade/default-probability (which is a PRE-TRADE gate on ONE position), nor of /v1/mcp/default-probability (which reads default EVENT COUNTS and a payment delay), nor of /v1/index/default-probability (which reads an INDEX CONSTITUENT SET), nor of /v1/counterparty/solvency-score (which reads the OBLIGOR'S OWN OBLIGATION LADDER). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.811Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"6d9c9a50-6d8d-4d6d-bd90-0ae10995e07b","method":"POST","path":"/v1/counterparty/exposure-score","description":"A deterministic COUNTERPARTY-EXPOSURE-SHAPED score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of the exposure you carry against one counterparty (date, counterparty_exposure_mark_to_market_usd, counterparty_exposure_collateral_usd, counterparty_exposure_haircut_pct, counterparty_exposure_liquidation_days, counterparty_exposure_rollover_count, counterparty_exposure_limit_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the EXPOSURE THE HOLDER CARRIES AGAINST ONE COUNTERPARTY -- its collateral cover after the haircut, its unwind time, its funding rollover dependency and its limit utilisation -- which no sibling does: the rubric weights counterparty_exposure_collateral_adequacy (0.30), counterparty_exposure_unwind_overhang (0.25), counterparty_exposure_rollover_dependency (0.25) and counterparty_exposure_limit_utilisation (0.20). It is NOT a rename of /v1/pretrade/exposure-score (which is a PRE-TRADE gate on ONE position), nor of /v1/index/exposure-score (which reads an INDEX CONSTITUENT SET), nor of /v1/counterparty/solvency-score (which reads the OBLIGOR'S OWN OBLIGATION LADDER). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.226Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f7d5b18b-96d6-4ee5-acb6-fb917f24fecf","method":"POST","path":"/v1/counterparty/score","description":"Price the risk of paying an unknown agent, endpoint or host. You supply YOUR OWN observations (settlements, disputes won/lost, uptime, first-seen, volume) and the engine scores them deterministically into 0-1000 with a band (prime / standard / watch / avoid), the weighted factors behind it, and a derived confidence. A counterparty with ZERO observations returns `insufficient` with the coverage gap named — never `prime`. At a fixed dispute rate, more settled volume never lowers the score. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.095Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8c1f1685-7fc7-4ccf-81ae-5805694a5a71","method":"POST","path":"/v1/counterparty/solvency-score","description":"A deterministic COUNTERPARTY-SHAPED SOLVENCY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of one obligor's obligation ladder (date, obligations_due_30d_usd, obligations_due_365d_usd, unencumbered_assets_usd, operating_cashflow_usd, secured_borrowings_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the OBLIGOR'S OWN OBLIGATION LADDER against its free (unencumbered) asset base and its cash generation, which no sibling does: the rubric weights obligor_near_term_wall (0.30), obligor_refinancing_dependency (0.25), obligor_asset_encumbrance (0.25) and obligor_earnings_stability (0.20). It is NOT a rename of /v1/counterparty/score (which reads BEHAVIOUR), nor of /v1/mcp/solvency-score (which reads balance-sheet RATIOS), nor of /v1/index/solvency-score (which reads an INDEX CONSTITUENT SET). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.524Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f82506e2-e015-4f40-9b77-78d6447bd810","method":"POST","path":"/v1/credentials/issue","description":"Issue a SCOPED, EXPIRING CAPABILITY CREDENTIAL for an agent -- the missing half of the permissions idea. The response is a compact signed credential {credential_id, subject, scopes[], issued_at, expires_at, max_uses, policy_sha256, signature, key_id} that ANY verifier can check OFFLINE. ONE TOKEN FORMAT, NOT TWO: the credential uses the SAME envelope the prepaid-capacity pass established (ps1.<b64url(payload)>.<b64url(ed25519_signature)>), signed over credential_id|kind|subject|scopes_csv|issued_at|expires_at|max_uses -- a verifier that understands a pass token understands a credential. SCOPES ARE VALIDATED AGAINST A DOCUMENTED VOCABULARY: an unknown scope is a 400 with the vocabulary in the body, and a credential can NEVER carry an unverifiable wildcard ('*', 'all', 'admin' are refused). The expiry is a SIGNED claim, so the bearer cannot move it. Verification is READ-ONLY and does not consume a use. A credential authorises a CALL, not a transfer of value. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.917Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3628e0b5-fa54-469c-bab7-f83a5506c9fa","method":"POST","path":"/v1/data/c2pa/verify","description":"Pay-per-verify C2PA content-provenance attestation. Caller supplies a content hash (64-char lowercase hex, 0x prefix optional) + a C2PA-style manifest claim; the deterministic pure-function engine (c2pa_kb.py) verifies manifest well-formedness, binds the content_hash against the manifest's declared asset hash, extracts creator/generator/software_agent/signature/timestamp/assertion/ingredient counts, checks structural integrity (manifest_version, has_signature, has_timestamp, content_binding_present), and grades the provenance into ONE Ed25519-signed, SHA-256-reproducible attestation keyed by attestation_id with verdict in {AUTHENTIC, SUSPICIOUS, MALFORMED} + an issue list. Optional expected_creator / expected_generator strings are compared verbatim. No live network call is made and nothing is fabricated — the same inputs re-analysed against the same anchor yield byte-exact the same attestation. Read back free at GET /v1/data/c2pa/attestation/{attestation_id}.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:10.795Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b79e8787-f856-4caf-ad8f-007442525a67","method":"POST","path":"/v1/data/conflict-resolution","description":"Conflicting facts, and a REASONED resolution. Detects the contradiction, weights the CALLER's declared source qualities, applies timestamp precedence, and reports WHICH factor decided it. THE RULE: when the supplied records do not decide the question, the verdict is `unresolved` and the disagreement is PRESERVED — both claims, their scores and the specific reason no winner was selected. It never invents a consensus. Numbers compare by VALUE, not by text, and claims about different subjects are not in conflict. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.010Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"58a8b7c0-d012-4e69-95ef-30d607a5d24b","method":"POST","path":"/v1/data/consent-chain","description":"Walk a consent chain and report whether the use is permitted at EVERY hop. Grades five dimensions per hop — consent_exists, scope_covers, party_covered, conditions, not_revoked — as permitted / not_permitted / unclear / unverifiable, and names the FIRST breaking hop. THE RULE: a chain is permitted only if EVERY hop is — a break at hop 3 invalidates the whole chain, and a hop that could not be read is NEVER reported as permitted. An empty chain is `unverifiable`: 'nobody objected' and 'everybody consented' are different claims. This reads the chain the caller supplied; it fetched no consent record, register of processors or regulatory database. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.730Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b849bf8a-54bd-477b-bd0b-b6fe4418e13d","method":"POST","path":"/v1/data/deletion-proof","description":"Prove a data subject's deletion was actually executed, from the evidence the caller supplies. Grades each in-scope target evidenced / not_evidenced / partial / unverifiable / contradicted, and returns the RESIDUE — every target the supplied receipts do NOT cover. THE RULE: a receipt that asserts deletion without carrying a digest, signature, reference or record count is `unverifiable`, NEVER `evidenced` — and a scope that names no target is `unverifiable`, never a clean bill. The residue is the product: a proof that reports only what was deleted, and is silent about what was not, certifies nothing. This service connected to NO database, backup, warehouse or processor. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.834Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"4c1d82e4-77bb-4083-bfee-e4e0728561c2","method":"POST","path":"/v1/data/freshness","description":"Answer 'is this data still good?' for a payload, document or claim. Verdict is fresh | stale | undated | conflicting. NO FABRICATION: a datum with no usable as_of is UNDATED and is_stale=true — absence of evidence of staleness is NOT evidence of freshness, and it is never reported as fresh. A FUTURE as_of is an INVALID input and is rejected with HTTP 400 invalid_timestamp — a datum from the future is not 'very fresh', it is a broken clock. CONFLICTS: when two or more supplied claims give DIFFERENT values for the same key, the verdict is 'conflicting' and BOTH values are reported — the engine never picks one, never takes the majority and never takes the newest. An optional source_url may be corroborated by an observed HTTP Last-Modified/Date header, reported SEPARATELY under observed_http and never overwriting the claimed timestamp; a network failure becomes a coverage_gap and the response is still valid, never a 500. Deterministic; no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.952Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"a9e055c6-3053-4600-8eb6-f8d5af57d3b0","method":"POST","path":"/v1/data/jurisdiction-resolve","description":"Which jurisdiction's rules apply to this data flow, given the caller's facts? Applies a PUBLISHED rules set — nine rules covering GDPR, the UK GDPR, the CCPA/CPRA, the LGPD, cross-border transfer mechanisms and processing locality — and reports each as applies / may_apply / does_not_apply / unclear with the fact it rests on quoted back. THE RULES SET IS SERVED IN FULL in the response so every attachment can be challenged. THE RULE: a rule whose trigger fact was not supplied is `unclear` and is NAMED — never silently dropped, never reported as applying, and the verdict is `partially_resolved` while any rule is unresolved. This service fetched no statute, case, guidance, transfer register or adequacy decision; it is a deterministic classifier over a published table, not legal advice. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.112Z","variable_amount":false,"min_price_usd":1.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1250000","price_usd":"1.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"a8722c29-5739-42f2-bc1c-a605ece47db6","method":"POST","path":"/v1/data/signed-snapshot","description":"Seal caller-supplied series rows + source citations into ONE signed, SHA-256-reproducible data snapshot with honestly-labelled freshness evidence (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.021Z","variable_amount":false,"min_price_usd":0.3,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"300000","price_usd":"0.3000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"51d40582-2826-4917-9522-7c0622639579","method":"POST","path":"/v1/data/source-independence","description":"Do these N 'sources' actually share one origin? Builds the source-dependency graph from the claims the caller supplies and returns independent_corroboration_count — the number of CONNECTED COMPONENTS once the dependency edges are removed. Four signals, each reported separately: shared reference, token-shingle Jaccard text similarity, identical timestamp, and a stated common upstream. THE RULE: an identical timestamp is a WEAK signal and never creates a dependency edge on its own — a wall of sources published in the same second is a fact about publishing, not about independence. Text similarity is a declared shingle size and a declared threshold, not vibes. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.528Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"26806140-8057-4fc3-878a-a07b649b8764","method":"POST","path":"/v1/data/usage-rights","description":"What does this licence permit, DIMENSION BY DIMENSION? Returns permitted | prohibited | conditional | unclear for commercial, redistribution, derivatives, attribution and model_training in ONE call, with the clause quoted and the condition named. Reads a supplied SPDX identifier OR a supplied custom licence TEXT. THE RULE: a dimension no supplied clause addresses is `unclear`, NEVER `permitted` — silence in a licence is not permission, and an unknown identifier with no supplied text settles nothing. A sentence that both grants and restricts resolves to the RESTRICTION. THIS IS AUTOMATED LICENCE INTERPRETATION, NOT LEGAL ADVICE. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.906Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e1d7959c-4823-4966-8e5e-c3a855df6978","method":"POST","path":"/v1/decision/attest","description":"Sign ONE decision record: what was decided, by whom, under which policy, on what evidence, at which assurance level (A1_UNATTESTED / A2_EVIDENCE_BOUND / A3_COUNTERSIGNED / A4_REPRODUCIBLE), and mandatorily what was NOT known at decision time. Signed evidence is re-hashed and checked; a tampered item caps the assurance. The engine never asserts truth — only what was signed, what was checked and what was missing. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.217Z","variable_amount":false,"min_price_usd":2.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2500000","price_usd":"2.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ece1321f-2f5b-4422-b2d9-448189524f49","method":"POST","path":"/v1/delivery/attest","description":"Proof that a digital deliverable actually arrived and matched what was promised. Returns delivered | mismatch | unacknowledged | insufficient_evidence, with hash_match, byte_match and a step-by-step chain. THE RULE: an absent promised_sha256 yields hash_match: null and the verdict CANNOT EXCEED `unacknowledged` — we will not certify `delivered` against a promise we never saw. A digest mismatch is `mismatch`, and it is never softened to `unacknowledged`. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.604Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"34e4f1b3-4171-4a5b-a2ab-240b4ea43e5a","method":"POST","path":"/v1/dispute/ruling","description":"ADVISORY ADJUDICATION on a claimed failure of delivery -- a signed ruling produced by a DETERMINISTIC RUBRIC over the evidence the parties supply. NOT arbitration, NOT legal advice, NOT binding on anyone, and we never move funds. THE RUBRIC (five tests, each PASS | FAIL | NOT_ESTABLISHED): artifact_present, manifest_signature_valid, coverage_window_met, quantity_satisfied, receipt_present. OUTCOMES: fulfilled | partial | unfulfilled | INCONCLUSIVE. 'inconclusive' is a FIRST-CLASS outcome and must be reachable: insufficient evidence never defaults to a ruling in anybody's favour -- a NOT_ESTABLISHED required test forces 'inconclusive' and a 50/50 split. PUBLISHED SPLIT FORMULA: delivery_ratio = min(observed_rows/required_rows, observed_days/required_days), each capped at 1.0; split_pct.seller = 100 * delivery_ratio; split_pct.buyer = 100 * (1 - delivery_ratio); fulfilled -> 0/100; unfulfilled -> 100/0; inconclusive -> 50/50. The formula is stated in the response itself. Same inputs => the same output_sha256, always. No LLM, no sentiment, no judgement calls. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.205Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d29ed125-299b-440e-8759-20be90a58a42","method":"POST","path":"/v1/edge/prior","description":"THE PRIOR-TEST REGISTRY -- has this been tested before, and what was the result? INCLUDES THE FAILURES. Public quant libraries sell what worked; nobody publishes negative results with sample sizes. That is the moat: a negative result cannot be back-filled because it was never published anywhere, and the registry grows every month. REQUEST: {instrument?, timeframe?, strategy_type?, family?, free_text?} -- every field optional, but at least ONE is required or the route answers 400 with the accepted field list. Instruments are NORMALISED, so USDJPY, USD/JPY and usd-jpy are one instrument; timeframes likewise (60m == H1). MATCH RANKING (published, deterministic): matches are sorted by the tuple (tier, instrument_rank, timeframe_rank, strategy_rank, family_rank, -text_overlap, name, id). tier 0 = exact instrument+timeframe; 1 = exact instrument, other timeframe; 2 = family or strategy_type; 3 = free-text token overlap. EXACT INSTRUMENT+TIMEFRAME ALWAYS RANKS ABOVE A FAMILY MATCH. THREE-STATE HONESTY: no match returns found=false, verdict.tested=false and coverage_gaps=['no_prior_test']. The service NEVER answers 'edge confirmed' and NEVER invents a grade. A CONTAMINATED result (lookahead-biased) is REPORTED but is NEVER counted as clean evidence -- if the only matches are contaminated, clean_evidence is false and the summary says so in words. OUTPUT WHITELIST: only aggregate fields are exposed -- no filesystem paths, no internal notes, no account references, no keys, no PII. Same query => same output_sha256, always. No LLM in the answer path. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:10.825Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8838bc5d-3036-4ee6-858a-4dbbbb09c7f6","method":"POST","path":"/v1/edge/replication","description":"IS THAT TRACK RECORD REAL? -- the high-ticket JUDGEMENT service. Anyone can compute a Sharpe ratio; this prices the question that decides whether you deploy real capital behind a backtest: given that you tried N things and kept the best, how much of this Sharpe would you have found BY CHANCE, and does any of it survive out of sample? REQUEST (all three REQUIRED): {returns: [floats] | returns_csv_b64: string, frequency: 'daily'|'per_trade', trials_attempted: int >= 2}; OPTIONAL: {claimed: {sharpe, cagr, max_dd}, strategy_shape: {type, instrument, timeframe}}. TRIALS_ATTEMPTED IS MANDATORY -- a Deflated Sharpe without the trial count is meaningless, and we NEVER assume trials = 1. A missing trial count returns verdict='insufficient_evidence' naming the field. A single trial is refused: the expected-maximum-Sharpe approximation is undefined at N=1 and would report maximum confidence for a one-configuration search. WHAT YOU GET: (1) a DEFLATED SHARPE (Bailey & Lopez de Prado 2014) against your own trial count, so a Sharpe of 2.0 found after 500 tries is priced for what it is; (2) a PBO (Bailey, Borwein, Lopez de Prado & Zhu 2015) -- the probability that the selection procedure picks a loser; (3) an out-of-sample Sharpe distribution CALIBRATED against a search-matched null, so 'it did well OOS' is measured against 'as well as a lucky search would have'; (4) an ANONYMISED base rate from our own graded vault: of the strategies of this shape we have actually graded, what fraction passed? Published ONLY at n>=5; below that it is omitted and the omission is named in coverage_gaps -- never interpolated, never a named strategy, never a path. FOUR STATES, NEVER THREE: replicable | likely_overfit | not_replicable | insufficient_evidence. PUBLISHED THRESHOLDS (every label is reproducible from these): pbo > 0.50 => high-severity finding 'pbo_above_threshold'; deflated_sharpe < 0.05 => not distinguishable from a lucky search; deflated_sharpe below the 0.95 significance level => the observed Sharpe is not significant once the search is accounted for; median OOS Sharpe excess over the search-matched null < 0 => not replicable; n < 100 (daily) or n < 30 (per_trade) => insufficient_evidence; zero variance => insufficient_evidence. Every threshold is echoed in assumptions.thresholds so a buyer can re-derive every label. DETERMINISTIC: no LLM in the answer path, no clock in the signed core, every split and null draw seeded -- the same request always produces the same output_sha256, byte for byte, in any process. LIMITS: input <= 1 MB and <= 200000 observations; beyond that the route answers 400 naming the limit. Verify the answer FREE at POST /v1/edge/replication/verify; the Ed25519 public key is at GET /v1/edge/replication/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.027Z","variable_amount":false,"min_price_usd":3,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"3000000","price_usd":"3.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"80a88554-1eb3-416b-bcb5-a5f5d1854fe4","method":"POST","path":"/v1/evidence/ingest","description":"Seal one signed-attribution / forensics evidence record ('who did what', actor DID + action hash + timestamps) into a tamper-evident vault entry (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.202Z","variable_amount":false,"min_price_usd":0.2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"200000","price_usd":"0.2000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e7ef6df3-9abb-4f53-a736-95c7439f1770","method":"POST","path":"/v1/evidence/query","description":"Bind a set of sealed forensics-vault entries (or an actor filter) into ONE signed, SHA-256-reproducible query receipt (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.035Z","variable_amount":false,"min_price_usd":0.15,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"150000","price_usd":"0.1500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"753d300e-6d6d-48de-87d2-ceda093c5fd4","method":"POST","path":"/v1/finance/attested-evidence","description":"Assess and attest a financial statement or metrics with signed, deterministic, honest evidence.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.553Z","variable_amount":false,"min_price_usd":0.35,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"350000","price_usd":"0.3500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f8df11dc-6083-41de-8ea1-83c4b6073dbd","method":"POST","path":"/v1/finance/conduct-score","description":"A deterministic FINANCE-LANE CONDUCT SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated FINANCIAL-STATEMENT observations of ONE counterparty or asset (date, finance_conduct_disclosure_defect_count, finance_conduct_warranty_breach_count, finance_conduct_warranty_breach_notional_usd, finance_conduct_disclosed_notional_usd, finance_conduct_filing_lag_days, finance_conduct_obligor_datum_present_count, finance_conduct_obligor_datum_expected_count) and the engine scores the CONDUCT EVIDENCE carried in that disclosure record under a published rubric onto an ABSOLUTE 0-100 score (higher is cleaner disclosure conduct) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights finance_conduct_disclosure_defect_ratio (0.30), finance_conduct_warranty_breach_ratio (0.25), finance_conduct_filing_lag_depth (0.20) and finance_conduct_obligor_disclosure_completeness (0.25). It is NOT a rename of /v1/mcp/conduct-score (the MCP/A2A interaction record), of /v1/audit/conduct-score (the audit lane's reading of observed conduct quality), of /v1/counterparty/default-probability (the fulfilment-performance record), of /v1/counterparty/exposure-score (the exposure the holder carries), of /v1/counterparty/solvency-score (the obligor's own obligation ladder), nor of /v1/audit/evidence-package (an evidence package, not a score). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative notional, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. The scale is ABSOLUTE, not peer-relative, so it can be used as a price or a limit input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.093Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"2cddcb5d-81f5-4452-81b9-8e6032b862f0","method":"POST","path":"/v1/finance/default-probability","description":"A deterministic FINANCE-LANE COUNTERPARTY/ASSET DEFAULT PROBABILITY SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated BALANCE-SHEET observations of ONE counterparty or asset (date, finance_default_probability_total_debt_to_equity_ratio, finance_default_probability_cash_and_equivalents_usd, finance_default_probability_operating_cash_flow_usd, finance_default_probability_interest_bearing_debt_usd), and the score reads FOUR things no sibling reads: how LEVERAGED the counterparty became, how FAST its liquidity runway depletes, how FAST its earnings convert to cash, and how DEPENDENT it is on rolling borrowed debt. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.333Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"a10256fe-0015-4c13-bb95-65104d589543","method":"POST","path":"/v1/finance/exposure-score","description":"A deterministic FINANCE-LANE COUNTERPARTY/ASSET EXPOSURE SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated EXPOSURE observations of ONE counterparty or asset (date, finance_exposure_observed_notional, finance_exposure_peak_notional, finance_exposure_encumbered_share, finance_exposure_realisable_notional, finance_exposure_aging_days) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is stronger standing) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights finance_exposure_concentration_ratio (0.25), finance_exposure_peak_intensity_ratio (0.30), finance_exposure_ageing_spread_ratio (0.20) and finance_exposure_encumbrance_ratio (0.25). Every component is non-increasing in risk and every weight is non-negative, so a strictly worse observation set can never score better -- and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check that the score does not go UP. The published rubric ships with the answer so you can verify that rather than trust it. The scale is ABSOLUTE, not peer-relative, so it can be used as a price or a limit input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.235Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"323586f4-b098-4b43-93fe-28cf29417329","method":"POST","path":"/v1/finance/issuer-score","description":"A deterministic FINANCE-LANE ISSUER STANDING SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated FILING and STANDING-DOCUMENT observations of ONE issuer (date, finance_issuer_filing_due_date, finance_issuer_filing_completed_date, finance_issuer_authorized_issuance_capacity, finance_issuer_committed_issuance_amount, finance_issuer_regulatory_action_severity, finance_issuer_standing_document_count, finance_issuer_standing_document_defect_count) and the engine scores whether the issuer has the LEGAL AND REGULATORY STANDING to issue the instrument it claims to issue, under a published rubric onto an ABSOLUTE 0-100 score (higher is stronger issuer standing) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights finance_issuer_filing_currency_deficit_ratio (0.30), finance_issuer_issuance_headroom_deficit_ratio (0.25), finance_issuer_regulatory_action_severity_intensity (0.25) and finance_issuer_standing_document_defect_ratio (0.20). It is NOT a rename of /v1/finance/conduct-score (the CONDUCT EVIDENCE carried in the disclosure record), of /v1/finance/solvency-score (the ability to MEET OBLIGATIONS), of /v1/mcp/issuer-score (registration and attestations), of /v1/audit/issuer-score (the audit lane's reading) nor of /v1/index/issuer-score (an index constituent set). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative magnitude, a defect count above the number of standing documents reviewed, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. The scale is ABSOLUTE, not peer-relative, so it can be used as a price or a limit input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.556Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"21d5b089-6fca-486c-afc1-a026a98a246d","method":"POST","path":"/v1/finance/risk-score","description":"A deterministic FINANCE-LANE COUNTERPARTY/ASSET RISK SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated OBSERVATIONS of ONE counterparty or asset (date, finance_risk_observed_value, finance_risk_adverse_event_count, finance_risk_settlement_delay_days, finance_risk_settlement_shortfall_ratio) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is stronger standing) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights finance_risk_observation_dispersion_ratio (0.20), finance_risk_adverse_event_intensity (0.30), finance_risk_settlement_delay_severity (0.25) and finance_risk_settlement_shortfall_ratio (0.25). It is NOT a rename of /v1/finance/conduct-score (the CONDUCT EVIDENCE carried in the disclosure record), of /v1/finance/solvency-score (the ability to MEET OBLIGATIONS), of /v1/finance/issuer-score (legal and regulatory standing), of /v1/mcp/risk-score (concentration, leverage, liquidity and loss events), of /v1/audit/risk-score (control deficiency, limit breach, model drift and policy exception) nor of /v1/index/risk-score (index volatility, drawdown, exit liquidity and constituent-move correlation). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative magnitude, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. The scale is ABSOLUTE, not peer-relative, so it can be used as a price or a limit input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.105Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ad0cca61-ed2b-45d9-a1d8-b5e39b3e366f","method":"POST","path":"/v1/finance/solvency-score","description":"A deterministic FINANCE-LANE SOLVENCY SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated FINANCIAL-PERIOD observations of ONE counterparty or asset (date, finance_solvency_period_current_liabilities_usd, finance_solvency_period_current_assets_usd, finance_solvency_period_interest_expense_usd, finance_solvency_period_ebitda_usd, finance_solvency_period_fixed_charge_usd, finance_solvency_period_net_sales_usd, finance_solvency_period_average_total_assets_usd) and the engine scores how WELL the counterparty can MEET ITS OBLIGATIONS under a published rubric onto an ABSOLUTE 0-100 score (higher is stronger solvency) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The rubric weights finance_solvency_working_capital_deficit_ratio (0.30), finance_solvency_fixed_charge_coverage_shortfall (0.25), finance_solvency_interest_coverage_erosion_ratio (0.25) and finance_solvency_asset_turnover_decline_ratio (0.20). It is NOT a rename of /v1/finance/conduct-score (the CONDUCT EVIDENCE carried in the disclosure record), of /v1/audit/solvency-score (the audit lane's reading), of /v1/counterparty/solvency-score (the obligor's own obligation ladder), of /v1/index/solvency-score (an index constituent set) nor of /v1/mcp/solvency-score (an MCP/A2A counterparty). The verdict is exactly one of: `pass` (the score is computed from every component the rubric needs and the components and their values are returned so the buyer can recompute it), `fail` (the observations violate the rubric's own consistency requirements - a negative magnitude, current assets above the period's own total assets, a duplicate dated observation - so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. The scale is ABSOLUTE, not peer-relative, so it can be used as a price or a limit input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.721Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"212a6481-9f61-4142-a3b1-1a7f0e3185e7","method":"POST","path":"/v1/forensics/decision-trail","description":"Hash-chain ONE multi-step decision trail for an agent run and sign it: every step's hash binds the previous step's hash, so a mutation anywhere in the middle of the trail is detected AT THE STEP IT OCCURRED, not merely as a digest mismatch at the end. Returns the per-step chain, the chain head, a computed tamper verdict (CHAIN_INTACT / CHAIN_BROKEN_AT_STEP / CHAIN_HEAD_MISMATCH / CHAIN_UNVERIFIED), the per-step expected-vs-actual diff (MATCH / MISMATCH / NOT_SUPPLIED / UNVERIFIED) and every named unresolved item. Integrity is not truth: a valid signature proves only that this record was signed. The Ed25519 public key, the canonicalisation rule and the signature are embedded in the response, so it verifies offline with no second call.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.265Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f1a2e134-0df4-4c9e-a1ea-76024df7bedb","method":"POST","path":"/v1/fund/tearsheet","description":"The institutional tearsheet, REPRODUCIBLE. Every ratio — CAGR, volatility, Sharpe, Sortino, max drawdown, Calmar, hit rate, skew, excess kurtosis, and the benchmark alpha/beta/correlation — is computed from the returns series you supply, and the FORMULA VERSION is published so a reader can re-derive every figure in a spreadsheet. Returns computed | insufficient_history | insufficient_input. Fewer than 24 periods is `insufficient_history` with the count named — a Sharpe on 10 periods is not a Sharpe. A `returns` value that is not a plain list of numbers is REJECTED, never coerced. A zero-variance series yields no Sharpe. Heavy tails are flagged. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.859Z","variable_amount":false,"min_price_usd":5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"5000000","price_usd":"5.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"84192880-b96f-41a1-8185-de0d301e87ac","method":"POST","path":"/v1/governance/reserve","description":"Agent Spend Governor — ATOMIC reserved lease. Caller supplies {agent_id, amount (USD), intent, ttl_seconds?}. The reservation is a single serialised DB transaction (BEGIN IMMEDIATE) that re-reads the live reserved total and REFUSES if a per-intent/session/daily cap would be exceeded — two concurrent reserves can never double-spend. Returns a signed, expiring lease (Ed25519, verifiable free). The adoption entry check route is FREE at POST /v1/governance/check.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-14T14:16:42.398Z","last_seen_at":"2026-11-19T01:40:13.296Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"6e594495-823a-4e47-8cf8-16fcb55d3888","method":"POST","path":"/v1/governance/settle","description":"Agent Spend Governor — settle a reserved lease with the actual amount spent and release the remainder. IDEMPOTENT on lease_id: replaying the same lease returns the recorded outcome and never double-counts spend. Settling an expired lease is explicit (documented), never a silent success. Caller supplies {lease_id, actual_amount (USD)}.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-14T14:16:42.398Z","last_seen_at":"2026-11-19T01:40:12.785Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d9ba7276-f16e-4fdb-9791-24a1d3ec644a","method":"POST","path":"/v1/identity/agent/attest","description":"Bind an agent to its operator and its wallet, with the evidence that supports the binding and an HONEST grade of that evidence. Returns attested | declared | insufficient_input. Evidence that cannot be verified is NEVER dropped and NEVER promoted — it is carried in `unverified` with the reason stated, because a gap that is hidden is a gap that gets relied upon. `verified` is deliberately UNREACHABLE: this engine does not observe a chain, so it will not assert third-party state. The binding_id is a DETERMINISTIC content hash, so the same binding always carries the same id — in this process and the next. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.129Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"bfb65b47-c673-4df5-9d08-9dfe6e52bd97","method":"POST","path":"/v1/identity/control-proof","description":"Attest, via a NON-DETERMINATIVE signed CONTROL-PROOF, that an entity documents CONTROL of an identifier (a key/DID/wallet), wrapped around the live veritrack-attest spine (Ed25519 + SHA-256). Deterministic pure-function engine (identity_kb.py): carries caller fields verbatim, rejects missing required fields, no lookahead/leaks/fabrication, no KYC / no identity determination (compliant boundary). Verify free at GET /v1/identity/control-proof/verify.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.052Z","variable_amount":false,"min_price_usd":0.2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"200000","price_usd":"0.2000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"658303a2-f7d5-4ece-8d55-8ddfb8c40b3a","method":"POST","path":"/v1/index/default-probability","description":"A deterministic INDEX-LEVEL DEFAULT PROBABILITY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of the constituent SET's default risk (date, index_constituents_tracked, index_constituents_defaulted, index_default_cluster_count, index_default_rate_recent_pct, index_default_rate_prior_pct, index_absorption_buffer_usd, index_default_notional_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the default risk the SET carries in AGGREGATE, how the defaults CLUSTER rather than arrive independently, how the default profile DEGRADES, and how much of a default the set can ABSORB without contagion -- which no single-counterparty paper check can see. The rubric weights index_set_default_burden (0.30), index_default_clustering (0.25), index_default_profile_degradation (0.25) and index_default_absorption_capacity (0.20). The verdict is exactly one of pass, fail or indeterminate: `fail` means the observations violate the rubric's own consistency requirements, so the score is NOT DEFINED and is withheld; `indeterminate` means there are too few observations, so the score is WITHHELD and the count is reported. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.595Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3b278ac7-52ed-414c-ac83-2c07dcceff41","method":"POST","path":"/v1/index/exposure-score","description":"A deterministic INDEX-LEVEL EXPOSURE PROFILE score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated index-level exposure observations (date, index_gross_exposure_usd, index_net_capital_usd, index_largest_position_share_pct, index_exit_unwind_days, index_exit_impact_bps) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the SHAPE of the risk taken on, which no sibling does: the rubric weights index_exposure_leverage (0.30), index_position_concentration (0.25), index_exit_unwind_horizon (0.25) and index_exit_impact_cost (0.20). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.162Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8992535c-45ec-4d74-bbb9-5f0c3feb56b8","method":"POST","path":"/v1/index/issuer-score","description":"A deterministic INDEX-LEVEL ISSUER-SET score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated index-level issuer-set observations (date, issuers_in_index, issuers_surviving, membership_churn_count, issuers_refinancing_window, issuers_defaulting, linked_default_issuers) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the CONSTITUENT SET's survival, membership stability, refinancing pressure and default contagion, which no sibling does: the rubric weights issuer_survival_retention (0.30), membership_turnover_stability (0.25), issuer_refinancing_wall (0.25) and default_contagion_breadth (0.20). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.038Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c371894d-a0f6-4967-ad72-402e930dfe1a","method":"POST","path":"/v1/index/risk-score","description":"A deterministic INDEX-LEVEL AGGREGATE RISK score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated index-level observations (date, index_level, index_peak_level, index_return_pct, constituent_move_correlation, index_bid_ask_spread_bps) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the INDEX'S OWN MARKET BEHAVIOUR, which no sibling does: the rubric weights index_drawdown_severity (0.30), index_return_volatility (0.25), constituent_move_correlation (0.25) and index_exit_liquidity (0.20). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.843Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"0155aa9c-aa01-4abe-a4b7-7264923471ee","method":"POST","path":"/v1/index/solvency-score","description":"A deterministic INDEX-FAMILY SOLVENCY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated index-level observations (date, index_assets_usd, index_obligations_usd, liquid_reserves_usd, maturities_due_90d, constituents_in_default) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the CONSTITUENT SET's balance-sheet capacity, which no sibling does: the rubric weights asset_coverage_ratio (0.30), reserve_ratio (0.25), coverage_headroom (0.25) and obligation_stack (0.20). The verdict is exactly one of: `pass` (computed and recomputable), `fail` (the observations violate the rubric's own consistency rules, so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations, so the score is WITHHELD and the count is reported). The score is MONOTONIC: a strictly worse observation set can never score better, and the published rubric ships with the answer so you can verify that rather than trust it. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.915Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"16565717-4426-4c18-bccf-266336cafbb0","method":"POST","path":"/v1/index/x402","description":"The Synthex Agentic-Commerce Index — a DATED, SIGNED index of the agentic economy itself, computed from our own observation corpus of x402 settlements. Nobody else can compute it: the corpus is ours. The answer publishes the rules that produced it — eligibility (measured volume, buyer count, uptime), the anti-whale exclusion (any service whose single-buyer share exceeds 0.90 is EXCLUDED, because in this market one buyer accounts for almost all of one service's flow), the weighting scheme and a hard 25% weight cap — so a reader can re-derive the marks. A corpus that cannot support the requested breadth returns `insufficient_coverage` with the counts; the index is NEVER padded. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.239Z","variable_amount":false,"min_price_usd":5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"5000000","price_usd":"5.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c9b83d0f-fbd7-4269-9b66-20a8953b1828","method":"POST","path":"/v1/intel/conduct-score","description":"A deterministic INTEL-LANE CONDUCT SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE intelligence source's OWN output (date, intel_conduct_score_row_unattributed_assertion_count, intel_conduct_score_row_single_sourced_assertion_count, intel_conduct_score_row_lapsed_assertion_count, intel_conduct_score_row_refuted_assertion_count), and the score reads how much of what the provider ASSERTS is worth believing on its own evidence: how many of its assertions carry NO named dated source, how many rest on NO independent second source, how many were current when published but are NO LONGER current at as_of, and how many are CONTRADICTED by a later dated source with no later dated resolution. THE MEASUREMENT UNIT IS INTELLIGENCE ASSERTIONS -- the four adverse quantities are COUNTS OF ASSERTIONS, not ratios, not notionals, not durations and not probability masses, so no sibling that reads a ratio, a notional, a day count or a percentage point of probability can express this measurement. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the audit conduct score reads the AUDITOR's own sample of a counterparty, the finance conduct score reads the FILINGS, the mcp conduct score reads the MCP lane's tool interactions and the verify conduct score reads the RECONCILIATION of the counterparty's own record. A provider that states everything beautifully, attributes it to nobody, corroborates it with nobody and has been contradicted on it scores BADLY -- which no sibling can express. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.388Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3b5bbc17-08b5-4f0b-9654-8612af1ab42f","method":"POST","path":"/v1/intel/counterparty-score","description":"A deterministic INTEL-LANE COUNTERPARTY INTELLIGENCE SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of the intelligence attestations published ABOUT ONE counterparty (date, intel_counterparty_score_row_unattributed_attestation_count, intel_counterparty_score_row_uncorroborated_attestation_count, intel_counterparty_score_row_stale_attestation_count, intel_counterparty_score_row_contradicted_attestation_count), and the score reads how WELL CORROBORATED that counterparty's intelligence record is: how many of the attestations about it carry NO named dated source, how many rest on NO independent second source, how many were current when published but are NO LONGER current at as_of, and how many are CONTRADICTED by a later dated source with no later dated resolution. THE MEASUREMENT UNIT IS INTELLIGENCE ATTESTATIONS ABOUT A COUNTERPARTY -- the four adverse quantities are COUNTS OF DATED ATTESTATIONS, not ratios, not notionals, not durations and not probability masses, so no sibling that reads a ratio, a notional, a day count or a percentage point of probability can express this measurement. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the A2A counterparty score reads ANOTHER AGENT's transaction record, the AUDIT counterparty score reads the AUDITOR's own remediation record, the PRE-TRADE counterparty score gates ONE position and the VERIFY counterparty score reconciles the counterparty's OWN reported record. The intel-lane CONDUCT sibling scores the intelligence SOURCE's own output rather than the counterparty the intelligence is about. A counterparty the world asserts a great deal about, attributes it to nobody, corroborates it with nobody and has been contradicted on it scores BADLY -- which no sibling can express. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.869Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"6e3ccd60-d963-4e76-a0e0-98cef44b0895","method":"POST","path":"/v1/intel/solvency-score","description":"A deterministic INTEL-LANE SOLVENCY INTELLIGENCE SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of the intelligence attestations published ABOUT ONE counterparty's SOLVENCY COVERAGE (date, intel_solvency_score_row_uncovered_attestation_count, intel_solvency_score_row_single_source_attestation_count, intel_solvency_score_row_expired_attestation_count, intel_solvency_score_row_retracted_attestation_count), and the score reads how WELL ATTESTED that counterparty's solvency is: how many of the attestations about its solvency coverage carry NO named dated source, how many rest on NO independent second source, how many have EXPIRED (current when published but no longer current at as_of), and how many have been RETRACTED by a later dated source with no later dated replacement. THE MEASUREMENT UNIT IS INTELLIGENCE ATTESTATIONS ABOUT A COUNTERPARTY'S SOLVENCY COVERAGE -- the four adverse quantities are COUNTS OF DATED INTELLIGENCE ATTESTATIONS, not ratios, not notionals, not durations and not probability masses, so no sibling that reads a ratio, a notional, a day count or a percentage point of probability can express this measurement. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the MCP solvency score reads an AGENT's tool interactions, the AUDIT solvency score reads the AUDITOR's own dated sample, the INDEX solvency score reads an INDEX CONSTITUENT SET, the FINANCE solvency score reads the counterparty's FILINGS, the VERIFY solvency score reconciles the counterparty's OWN reported solvency record and the BALANCE-SHEET counterparty solvency score reads balance-sheet ratios. The intel-lane CONDUCT sibling scores the intelligence SOURCE's own output and the intel-lane COUNTERPARTY sibling scores the corroboration breadth of the intelligence about the counterparty, without reference to solvency. A counterparty whose solvency is asserted by a great many dated attestations that attribute it to nobody, corroborate it with nobody, have retracted it and have let it expire scores BADLY -- which no sibling can express. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.779Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cffa5c0f-487d-4d04-bcf3-91b981ed7f96","method":"POST","path":"/v1/invoice-ocr","description":"Extract structured fields (vendor, amount, date, line items) from an uploaded invoice or receipt.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.475Z","variable_amount":false,"min_price_usd":0.02,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"20000","price_usd":"0.0200000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"805c26f3-ed7a-4cc8-ac2f-5a4f07219263","method":"POST","path":"/v1/latency/benchmark","description":"Price $0.05 / 50000 atomic per benchmark. p50/p95/p99 latency for a third-party API, measured by us. Percentiles are computed ONLY from real samples; fewer than 20 samples and the verdict is 'insufficient_samples' with NO percentiles published. The answer states the probe method so a reader can reproduce it, and carries the error count beside the percentiles. Ed25519-signed and SHA-256 reproducible. Deterministic, no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.497Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"04f2f96e-7955-4fcb-8a58-1bac718ba085","method":"POST","path":"/v1/licence/check","description":"Data-Licence Compliance Oracle — a DETERMINISTIC verdict on whether a given use of a given dataset is permitted under a given licence. NO LLM in the answer path: a pure function over a curated licence corpus (MIT, Apache-2.0, BSD-3-Clause, GPL-2.0/3.0, AGPL-3.0, LGPL-2.1, MPL-2.0, CC0-1.0, CC-BY-4.0, CC-BY-SA-4.0, CC-BY-NC-4.0, ODbL-1.0, PDDL-1.0 and a proprietary-vendor-tos class). Returns verdict in {prohibited, permitted, conditional, unknown} with the VERBATIM clause, its source URL, the obligations, and the coverage gaps. SAFETY ASYMMETRY: an unsettled licence returns unknown and names the missing clause — it is impossible for an unsettled licence to return permitted. Ed25519-signed, SHA-256-reproducible; verify FREE at POST /v1/licence/verify, public key at GET /v1/licence/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.102Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e7a93b1f-bae1-4815-b2bc-664c1f7ac3f4","method":"POST","path":"/v1/mandate/compliance","description":"A signed, dated compliance certificate an allocator can file: does this portfolio comply with THIS mandate. Returns compliant / breach / indeterminate with every check reported (limit, actual, result), each breach graded by severity with a remediation, and a published methodology version. An asset class the mandate does not list, or a mandated metric you did not supply, returns `indeterminate` naming the gap — the engine never certifies compliance on missing evidence. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.275Z","variable_amount":false,"min_price_usd":5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"5000000","price_usd":"5.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"99e39e33-7c3b-45d1-8f8f-e46b54c0b923","method":"POST","path":"/v1/market/history","description":"PAID MARKET-DATA ARCHIVE — one bar SERIES over an explicit range, with an Ed25519-signed manifest. This is the RANGE route; POST /v1/market/snapshot is the TAIL route (the latest N bars of one instrument). The two never return the same thing: a snapshot request whose window starts more than 30 days back is rejected with HTTP 400 and a pointer at this route. REQUEST {source, instrument, timeframe, start, end, format}. source is 'binance' or 'ecb'. instrument is drawn from a TIGHT allowlist (no dynamic symbol fetching is performed — a symbol we do not hold is refused with the allowlist in the error body). timeframe is M1/M5/M15/H1/H4/D1 for binance and D1 ONLY for ecb (the ECB publishes daily reference rates only). format is 'json' (default) or 'csv'. CAPS: 1 instrument, <= 50,000 bars, range <= 2 years. Exceeding a cap returns HTTP 400 carrying the ACTUAL counts (requested_bars, limit) — never a silent truncation. RESPONSE {series, manifest}. The manifest carries exactly {source, licence, instrument, timeframe, start, end, rows, first_ts, last_ts, gaps, sha256, generated_at, key_id, signature}. sha256 is SHA-256 over the canonical serialisation of the series payload (stable JSON, sort_keys=true, separators=(',',':'), ensure_ascii=false) — recomputable by the buyer from the returned body alone. signature is Ed25519 over the preimage sha256_RAW_BYTES || generated_at_utf8 || rows_decimal_utf8. gaps is a TRUTHFUL gap report: the expected bar count for the window is compared against the observed rows and contiguous missing spans are listed as {from, to, bars_missing}. An empty gaps[] means VERIFIED COMPLETE, never 'not checked'. LICENCE: the data is the Binance official public bulk archive or the European Central Bank euro foreign-exchange reference rates, and every response and manifest carries that attribution. Restricted vendor FXTrade Rates are never served. Re-verify for free at POST /v1/market/history/verify; the Ed25519 public key is at GET /v1/market/history/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.075Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d2ec4bdb-6351-4e22-964f-75d5e7cdfe96","method":"POST","path":"/v1/market/history/bulk","description":"PAID MARKET-DATA ARCHIVE — BULK: up to 20 (instrument, timeframe) pairs in one call, <= 500,000 bars total, range <= 2 years. REQUEST {pairs: [{source, instrument, timeframe}, ...], start, end, format}. Same sources, allowlist, timeframe rules and manifest format as POST /v1/market/history. RESPONSE {results: [{instrument, timeframe, source, series, manifest}, ...], batch_manifest: {sha256, generated_at, key_id, signature, total_rows, results, partial_failures, sha256_scope}}. batch_manifest.sha256 is SHA-256 over the UTF-8 concatenation of each result's sha256 hex string IN REQUEST ORDER, and is signed the same way (preimage: sha256_RAW_BYTES || generated_at || rows). PARTIAL FAILURES: a pair that fails validation does NOT sink the batch — it is returned in batch_manifest.partial_failures with its reason, and the batch still settles, because the buyer pays for the work that WAS done. The batch is only refused (HTTP 422, nothing settles, no funds move) when NONE of the requested pairs could be served. A structural error (more than 20 pairs, a malformed date) is an HTTP 400 and also does not settle. LICENCE: Binance official public bulk archive or European Central Bank euro reference rates, attributed in every manifest. Restricted vendor FXTrade Rates are never served.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.271Z","variable_amount":false,"min_price_usd":5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"5000000","price_usd":"5.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ba853cc1-21ce-4514-a573-7dc85f6fd0a6","method":"POST","path":"/v1/market/regime","description":"Is this market trending or ranging? A signed, dated regime read (trend_up | trend_down | range | high_vol | low_vol | unknown) with a confidence, computed ONLY from the series you supply — the slope, the ATR percentage and the ADX are all reproducible from your own bars, with no external data and no OANDA. Fewer than 60 bars returns unknown with the count named. Three methods: trend, volatility, combined. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.191Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ef5394b9-b27f-4eb0-b450-1bcb7a6f303b","method":"POST","path":"/v1/mcp/conduct-score","description":"A deterministic MCP/A2A counterparty conduct score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, notional_usd, adverse, disputed) and the engine scores them under a published rubric into a 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative notional, a duplicate dated observation — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-increasing in risk and every weight is non-negative. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.692Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"52f1cddd-1a49-4087-a4f8-ebb79f230a17","method":"POST","path":"/v1/mcp/default-probability","description":"A deterministic MCP/A2A DEFAULT PROBABILITY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, default_events, payment_delay_days, adverse_events, dispersion) of a counterparty or an asset, and the engine scores the probability that the obligation DEFAULTS under a published rubric onto an ABSOLUTE 0-100 percent scale (higher = more likely to default), with the inverted 0-100 safety score, a probability band (negligible / low / moderate / elevated / high), every component and its raw inputs so you can RECOMPUTE the probability yourself, and a derived confidence. This is the GRADED question: the solvency score says whether a counterparty can meet its obligations, this one says how likely it is to miss the next one — which is what a pricing engine puts in a spread and a credit desk puts in a limit. The scale is ABSOLUTE, not peer-relative, so it does not move when the peer set does. The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative figure, a duplicate dated observation, a figure outside its published domain, or an impossible combination such as a default event with zero days past due — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few usable observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-decreasing in default risk and every weight is non-negative, and two of the four components are worst-case aggregates over the dated readings, so the guarantee holds for the whole series. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.026Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e20f2ea7-64b2-4ce5-bd69-a168082854f8","method":"POST","path":"/v1/mcp/eval/judge","description":"Grade ONE caller-supplied response pair (candidate vs reference/gold) against a rubric and return a SIGNED, SHA-256-reproducible LLM-judge grade (Ed25519 + SHA-256). Deterministic pure-function engine (judge_kb.py): no lookahead, no leaks, no fabricated data; a numeric grade is withheld when no rubric criterion is assessable. Verify the grade free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.276Z","variable_amount":false,"min_price_usd":0.15,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"150000","price_usd":"0.1500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"1efe49d2-b855-4d24-b94d-ebe02b22cfa2","method":"POST","path":"/v1/mcp/issuer-score","description":"A deterministic MCP/A2A ISSUER-LEGITIMACY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, registration_status, authority, good_standing, attestation_lapsed, attestation_withdrawn) of an entity that claims to issue an instrument, and the engine scores whether that entity IS A LEGITIMATE ISSUER — does it actually exist, is it in good standing, and does it have the AUTHORITY to issue the instrument it claims to issue — under a published rubric onto a 0-100 score (higher is more legitimate) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative figure, a duplicate dated observation, a numeric authority coverage outside 0..1, or an impossible combination — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few usable observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-decreasing in risk and every weight is non-negative, and three of the four components are worst-case aggregates over the dated readings, so the guarantee holds for the whole series. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.615Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8e5f9bf7-9d67-4952-8942-3d6a6bf64632","method":"POST","path":"/v1/mcp/risk-score","description":"A deterministic MCP/A2A RISK score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, concentration, liquidity_shortfall, leverage, loss_events) of an EXPOSURE — the position a desk is about to take, the trade it is about to make — and the engine scores how much risk that exposure carries under a published rubric onto a 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. This is the EXPOSURE score: the conduct score says whether a counterparty behaves well, the solvency score whether it can meet its obligations, the issuer score whether it is a legitimate issuer — this one says how risky the exposure to it is. The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative figure, a duplicate dated observation, a figure outside its published domain, or a self-contradictory pair such as loss events with zero liquidity shortfall — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few usable observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-decreasing in risk and every weight is non-negative, and three of the four components are worst-case aggregates over the dated readings, so the guarantee holds for the whole series. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.899Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"b053a65c-7d2f-433e-9d32-3881e05aee77","method":"POST","path":"/v1/mcp/solvency-score","description":"A deterministic MCP/A2A counterparty SOLVENCY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated financial observations (date, assets_usd, liabilities_usd, liquid_assets_usd, adverse) and the engine scores the counterparty's capacity to MEET ITS OBLIGATIONS under a published rubric into a 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative figure, liquid assets exceeding total assets, a duplicate dated observation — so the score is NOT DEFINED and is withheld), or `indeterminate` (too few usable observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-decreasing in risk and every weight is non-negative, and three of the four components are worst-case aggregates over the dated readings, so the guarantee holds for the whole series. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:10.998Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"76124143-898c-40df-aae7-dcc8cb0a59a6","method":"POST","path":"/v1/mcp/threatmodel/tool-permission","description":"Pay-per-analysis MCP that threat-models the caller's OWN MCP/A2A manifest for permission-sprawl takeover risk — the root of agent takeover. Caller supplies the manifest (MCP tools/list or normalized A2A skills) + (recommended) an ownership proof; the engine deterministically scores {overallRiskScore, riskClassHistogram, sprawlFindings[], verdict} into ONE Ed25519-signed, SHA-256-reproducible risk envelope. Auth-gated: CRITICAL verdicts require an ownership-proof'd manifest. Risk verdict only — no remediation authority. Verify free at POST /v1/mcp/threatmodel/tool-permission/recheck.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.247Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"aa166986-5d20-40b9-b237-5c837e523f62","method":"POST","path":"/v1/mcp/verify/uptime","description":"Pay-per-probe MCP that checks an x402 service is UP AND SETTLES. Caller supplies the probe checks + (recommended) an on-chain settlement anchor (lastConfirmedSettlementTx); the engine deterministically scores {up, lastConfirmedSettlementTx, settlementLatencyP95, checks[]} into ONE Ed25519-signed, SHA-256-reproducible uptime envelope. Settlement-anchor requirement is the differentiator on the crowded canary shelf (no live ping is made during the paid probe, so results are byte-exact reproducible). Verify free at POST /v1/mcp/verify/uptime/recheck.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.247Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c3764802-822c-46fd-bdf6-c9ae97d4e0cc","method":"POST","path":"/v1/memory/get","description":"Return the latest signed, SHA-256-reproducible sealed state record for an agent's memory key (verifiable cross-session continuity).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.653Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"68a0890d-743c-46c3-ace2-021a483479f7","method":"POST","path":"/v1/memory/put","description":"Persist one budget-capped state version into an agent's memory vault and append it to a hash-chained, Ed25519-signed continuity store (reproducible stateHash).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.007Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"0d657f0b-4ecb-43f9-981d-ba2049d22ad5","method":"POST","path":"/v1/nav/attest","description":"A signed NAV attestation for agent treasuries — the single document every allocator, auditor and counterparty asks for: what is this book actually worth, as of when, computed how. Returns nav_final / nav_provisional / nav_incomplete. A mark older than the staleness budget is flagged on the line and downgrades the verdict to `nav_provisional` — a stale mark is NEVER silently priced as current. A missing mark excludes the line, is named in `coverage_gaps` and downgrades the verdict to `nav_incomplete` — it is NEVER carried at zero. FX conversion names its source (the ECB euro reference rates) and publishes the rate and its date. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.085Z","variable_amount":false,"min_price_usd":10,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"10000000","price_usd":"10.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"5eccec16-291d-4aff-a39a-9a58511299ff","method":"POST","path":"/v1/operations/behaviour-drift","description":"The API is still up — but is it behaving the same? Stores each observation against the PERSISTED history for its (endpoint, observer) pair and diffs it against the previous one: fields added or removed, type changes, numeric value shifts, latency shift and price change. THE RULE: the FIRST observation for a pair returns baseline_recorded true with drift UNKNOWN — never 'no drift'. There is nothing to have drifted from yet, and 'no drift' against no baseline is a statement about nothing. The observation is the CALLER's; this service did not call the endpoint. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.104Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e57de402-758c-48c0-9ae7-a2684f8e6d1b","method":"POST","path":"/v1/operations/reproducibility","description":"Can THIS artifact be recreated from ITS inputs? Compares a recorded production context against a reproduction context — each carrying inputs, versions and environment — and reports WHERE they diverge: the input that differs, the version that differs, the environment variable that differs. THE RULE: a PARTIAL comparison is `insufficient_information`, never `reproducible`. This engine compared two contexts the CALLER supplied; it did not re-execute anything, did not run the caller's code, and did not confirm either context produced the artifact. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.667Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"1af219cc-34a3-4c20-8277-185da50fc86e","method":"POST","path":"/v1/output/verify","description":"Agent Output Verification Engine — independent QC verdict on a third-party agent output. Caller supplies the verbatim output + optional caller-supplied schema, fetchedAt anchor, sources, and accuracy reference; the deterministic pure-function engine (output_kb.py) grades FIVE legs (schema, freshness, consistency, evidence, accuracy) and folds them into ONE Ed25519-signed, SHA-256-reproducible verdict in {PASS, PASS-WITH-CAVEATS, FAIL, INSUFFICIENT-DATA}. NO LLM inference, NO live network probe, NO invented score — if fewer than 2 of 5 legs are assessable the verdict is honestly withheld as INSUFFICIENT-DATA. coverage_gaps name every leg honestly not assessed. Read back free at GET /v1/output/attestation/{attestation_id}; verify signature FREE at GET /v1/output/verify; standard at GET /v1/output/standard.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.715Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"92963687-ecfd-4f9a-bd39-3207fb9b96b9","method":"POST","path":"/v1/pass/purchase","description":"Prepaid Capacity Pass — buy 1,000 calls for $25.00, valid 30 days, covering any live paid route priced at or below $0.25. An effective $0.025/call: cheaper per call than paying individually, and committed spend paid up front instead of a micro-payment negotiated on every call. The response carries a COMPACT SELF-VERIFYING pass_token (Ed25519 over pass_id|route_scope|calls_total|expires_at) which the buyer's agent replays on each call via the X-Zeux-Pass header. The balance is decremented ATOMICALLY; the same Idempotency-Key on the same pass never double-decrements. A pass authorises micro-tier routes only — a $2.50 route returns 402 pass_scope. Verify FREE (read-only, consumes nothing) at POST /v1/pass/verify; public key at GET /v1/pass/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.206Z","variable_amount":true,"min_price_usd":null,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"25000000","price_usd":"25.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"f55fc0fc-98f9-4f31-82d2-9f13c17ae90c","method":"POST","path":"/v1/pass/purchase/trial","description":"Prepaid Capacity Pass — TRIAL TIER. $0.50 buys 20 calls, valid 7 days, covering any live paid route priced at or below $0.25 (an effective $0.025/call). This is the SAME pass machinery as the $25 tier: the same store, the same COMPACT SELF-VERIFYING pass_token format (ps1.<b64url(payload)>.<b64url(ed25519_signature)>, Ed25519 over pass_id|route_scope|calls_total|expires_at), the same scope 'micro_le_0.25', the same X-Zeux-Pass header and the same ATOMIC decrement with the same Idempotency-Key protection — ONE token format, not two, and NO new gate logic. A trial pass authorises micro-tier routes only: a $2.00 route returns 402 pass_scope. The $25 tier (1,000 calls, 30 days) is unchanged. Verify FREE (read-only, consumes nothing) at POST /v1/pass/verify; public key at GET /v1/pass/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.300Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e92a490b-1105-47b3-ba79-d64d2cc7833d","method":"POST","path":"/v1/payment/deadline","description":"Price $0.05 / 50000 atomic per verdict. Will this payment still settle inside my deadline? Judged ONLY on the MEASURED settlement times for the NAMED rail — we never extrapolate from another rail. Fewer than 20 measured settlements on that rail and the verdict is 'unknown' with the sample count, never a prediction. Otherwise: on_time (p95 fits) | at_risk (p50 fits, p95 does not) | late (even the p50 does not). Returns the slack, the p50/p95 basis and the measured_at instant, signed. Deterministic, no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.403Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"9c13d696-3076-4974-b73f-d5381f1b0310","method":"POST","path":"/v1/payment/envelope/validate","description":"Per-counterparty x402 conformance opcode. Submit ANY captured 402 payment-required envelope and receive a signed CONFORMANCE VERDICT over EIGHT named checks: accepts[] present, payTo recoverable (0x 20-byte EVM address), atomic price sane, CAIP-2 network declared, scheme declared, HTTPS resource, envelope fresh, x402 version declared — so an agent never pays an unpayable or non-settling endpoint. Verdict is CONFORMANT | NON-CONFORMANT with per-check observed values. A check that cannot run is reported not_observed in coverage_gaps, never as a pass. Static conformance read only: no network call, no chain resolution, no settlement. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.017Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d20b82e7-5183-4a34-8720-31b0f6b83859","method":"POST","path":"/v1/policy/check","description":"Evaluate a proposed commercial action against a CALLER-SUPPLIED policy set. Decision is allow | deny | needs_approval. FAIL-CLOSED: a missing required action field is a DENY with reason 'insufficient_action_detail', and an UNKNOWN rule key is reported as 'unsupported_rule' and FORCES needs_approval — silently ignoring a rule the buyer wrote is the one failure that would make this service dangerous, so it cannot happen. SUPPORTED RULE VOCABULARY: max_amount_usd, allowed_counterparties, blocked_jurisdictions, allowed_hours_utc (two entries = an inclusive UTC window, one = that hour, more = a discrete set; a window may wrap midnight), require_approval_above_usd, allowed_resources (exact string or a trailing-'*' wildcard), max_calls_per_hour (enforced from the CALLER's own supplied state — this engine has no call history and never pretends to). A rule that cannot be enforced is reported 'not_evaluated' and also forces needs_approval. Deterministic rule engine; no LLM; no registry lookup. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.086Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"548c270e-9481-463b-8a63-f2505a6400e4","method":"POST","path":"/v1/portfolio/risk","description":"Sign ONE exposure report computed ONLY from the caller's own positions and its own stated risk model: gross and net exposure, per-instrument concentration, effective number of positions, marginal risk contribution, a deterministic risk grade (R1_LOW / R2_MODERATE / R3_ELEVATED / R4_CRITICAL) and the NAMED assumptions the report required. A missing vol is reported as vol_assumed: none-supplied and degrades the grade — no vol is ever fetched or invented. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.505Z","variable_amount":false,"min_price_usd":3,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"3000000","price_usd":"3.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"19511156-e981-4f3e-a277-8c75c30b0727","method":"POST","path":"/v1/pretrade/counterparty-score","description":"A deterministic PRETRADE counterparty score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, disputes_raised, settlements_late, exposure_usd) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the PRE-TRADE SETTLEMENT record, which no sibling does: the rubric weights dispute_history (0.30), settlement_lateness (0.25), concentration_of_exposure (0.25) and relationship_tenure (0.20). The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative exposure, a duplicate dated observation — so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-increasing in risk and every weight is non-negative. The scale is ABSOLUTE, not peer-relative, so it does not move when the peer set does. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.878Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ba68bb15-42e3-44dd-a2e8-78f3647b891a","method":"POST","path":"/v1/pretrade/default-probability","description":"A deterministic PRETRADE DEFAULT PROBABILITY score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, days_to_maturity, open_obligations, cure_expectancy_pct, delinquency_flag) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the PRE-TRADE OBLIGATION record, which no sibling does: the rubric weights time_to_default (0.30), obligation_load (0.25), cure_expectancy (0.25) and delinquency_prevalence (0.20). The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative time to maturity, a duplicate dated observation — so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-increasing in risk and every weight is non-negative. The scale is ABSOLUTE, not peer-relative, so it does not move when the peer set does. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.892Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"851c0b62-cc1a-42fe-9ed4-4c7107e50636","method":"POST","path":"/v1/pretrade/exposure-score","description":"A deterministic PRETRADE EXPOSURE score with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations (date, gross_exposure_usd, limit_usd, adverse_excursion_pct, days_outstanding, risk_factor) and the engine scores them under a published rubric onto an ABSOLUTE 0-100 score (higher is safer) with a risk band (low / moderate / guarded / elevated), every component and its raw inputs so you can RECOMPUTE the score yourself, and a derived confidence. It reads the PRE-TRADE EXPOSURE record, which no sibling does: the rubric weights exposure_ratio (0.30), exposure_excursion (0.25), exposure_persistence (0.20) and exposure_diversification (0.25). The verdict is exactly one of: `pass` (scored, components returned), `fail` (the observations violate the rubric's own consistency rules — a negative notional, a duplicate dated observation — so the score is NOT DEFINED and is withheld with every violation named), or `indeterminate` (too few observations to score on, so the score is WITHHELD and the count is reported — never estimated). A strictly worse observation set can never score better: every component is non-increasing in risk and every weight is non-negative. The scale is ABSOLUTE, not peer-relative, so it does not move when the peer set does. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.887Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"25e8764b-2fe6-447a-aa24-e5bb2c6017e9","method":"POST","path":"/v1/pretrade/risk","description":"The order gate. Call it BEFORE every order: it decides approve / reduce / reject against the risk limits YOU declared, and a `reduce` returns a size that actually passes every check, not just a label. Six named checks — per-trade risk vs the stop distance, gross exposure, open-position count, instrument weight, correlated exposure against a static versioned bucket table, and margin headroom. An input too incomplete to decide returns `insufficient_input` naming the missing field, never a guessed verdict. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.596Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ee70d883-0090-404c-bc67-9fdc1846a50e","method":"POST","path":"/v1/procurement/reconcile","description":"Reconcile three ledgers — what was ORDERED, what was DELIVERED, what was PAID — line by line. Returns reconciled | variance | insufficient_evidence. Every variance names its DIRECTION (delivered_short, paid_over, and so on), because a variance that does not say which way it is sends the reader back to the spreadsheet. An ABSENT delivery or payment record is `insufficient_evidence`, NEVER a zero fill — a zero fill invents a shortfall the vendor may dispute forever. The amount ordered is COMPUTED from qty x unit_price, and the tolerance applied is reported. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.461Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"62a6b46d-f9f8-4dd5-a619-75168c012d03","method":"POST","path":"/v1/provenance/overlay","description":"Signed provenance for a piece of content, so a buyer can see where it came from. Returns provenance_attested | partially_attested | unattested, with every source and its hash_verified flag. THE RULE: a source we cannot verify is hash_verified: false and is NEVER DROPPED — dropping it would make the overlay look cleaner than the evidence supports, which is the one thing a provenance overlay must never do. No sources is `unattested`, NOT an error. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.150Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cd241848-dec5-4338-b7e4-2405d89fb6ff","method":"POST","path":"/v1/quant/monte-carlo","description":"What does the equity curve actually look like? A signed, dated distribution of terminal outcomes from a seeded IID resimulation: the p5/p25/p50/p75/p95 percentiles, the probability of profit, and the p50 and p95 maximum drawdown. The same seed yields the same output_sha256 in any process, so the digest is an identifier rather than a timestamp. More than 20000 runs is refused with the cap named. The method note states plainly that IID resampling ignores autocorrelation and understates the tail. Deterministic, signed.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.611Z","variable_amount":false,"min_price_usd":0.15,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"150000","price_usd":"0.1500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e323c898-6779-4c62-a67c-d9398285dda7","method":"POST","path":"/v1/quant/overlap","description":"Are these two strategies the same bet? A signed, dated correlation MATRIX, the crowded pairs above the threshold, and the CLUSTERS of strategies that move together. A pair with fewer than 60 overlapping observations is OMITTED and COUNTED as insufficient_pairs — never zero-filled, because a zero correlation is a measurement and a missing one is not. The verdict is independent | partially_correlated | crowded | insufficient_data. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.768Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"432079fc-3885-4830-9034-a7ce796e6012","method":"POST","path":"/v1/quant/risk-of-ruin","description":"Can this sizing scheme bankrupt me? A signed, dated ruin answer (safe | cautious | excessive | ruin_dominant | insufficient_input) from the classical FIRST-PASSAGE ruin model for a fixed-stake bettor (Dubins & Savage), with the Kelly and half-Kelly fractions beside it. RUIN IS ABSORBING: once equity reaches zero it stays zero. The analytic value AND a seeded simulation of the same process are both reported, and if they disagree by more than 0.05 the response says so rather than presenting one as the truth. A win rate outside (0,1) or a non-positive loss returns insufficient_input. Deterministic, signed.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.923Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c3fa2c9d-36c7-4b08-8c22-696be499d747","method":"POST","path":"/v1/quant/significance","description":"Is this return stream distinguishable from noise? A signed, dated one-sided SIGN-FLIP PERMUTATION test on an order-dependent statistic (the information ratio) — a test on a permutation-invariant quantity is vacuous, and this one is not. The p-value is corrected for the number of trials the caller states. A strongly NEGATIVE mean is not an edge and does not pass. The permutation null is seeded, so the p-value is reproducible across processes. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.892Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"13cb6df3-be26-49d0-a5d2-25b112416068","method":"POST","path":"/v1/readiness/attest","description":"Attest, from caller-supplied readiness evidence per a supplied policy snapshot, that a system/component/service is containment-ready. Wraps the live veritrack-attest spine (Ed25519 + SHA-256). Deterministic pure-function engine (readiness_kb.py): carries caller fields verbatim, rejects missing required fields, no lookahead/leaks/fabrication, NO kill-switch execution authority (compliant boundary). Verify free at GET /v1/readiness/attest/verify.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.044Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d24fc822-1c91-48d9-8f1a-9466fe2e73c8","method":"POST","path":"/v1/receipts/mint","description":"Mint a TYPED, signed attestation receipt for any tx/action — a structured per-transaction receipt schema distinct from flat notarization, wrapped around the live veritrack-attest spine (Ed25519 + SHA-256). Deterministic pure-function engine (receipt_kb.py): carries caller fields verbatim, rejects missing required fields, no lookahead/leaks/fabrication. Verify free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.449Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"95f926fa-d650-4f41-ab33-0958ca2d36f8","method":"POST","path":"/v1/research/attested-brief","description":"Produce a signed, verifiable research brief on a requested topic with reproducible source evidence.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.995Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d5c6af2b-078b-46ce-b283-896cb0b37d48","method":"POST","path":"/v1/route/execute","description":"An ADVISORY execution plan for moving money. This service NEVER moves money: it holds no key, signs nothing, broadcasts nothing, reads no credential and touches no ledger — it returns a PLAN, and the boundary is asserted against the module's own source. Returns executable | not_executable | no_viable_route | insufficient_input. A plan that cannot meet a constraint is `executable: false` with the BLOCKER named and by how much. Every fee comes from the declared rail table with its source. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.399Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"17fb2cde-db50-4ef4-9913-62513bf1ee1a","method":"POST","path":"/v1/sec-extract","description":"Extract structured company and SEC filing data for a stock ticker.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.304Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"3325d53d-7318-46c3-b183-dcbea6b563ae","method":"POST","path":"/v1/security/agent-scan","description":"The AGGREGATE agent assessment: run every detector whose input is present, fold the findings through the internal orchestrator, and return ONE signed report with a blocking verdict. A critical or high finding BLOCKS the report — a high aggregate score NEVER overrides a failure. A detector whose input is absent does not run, and the report NAMES it in `steps_not_run`. This is a security assessment, NOT a guarantee that the target is safe. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.484Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ec7425ac-3796-4edd-96f7-bd9530a574ef","method":"POST","path":"/v1/security/dependency-risk","description":"Assess and pin the dependency and supply-chain risk of a software dependency tree with signed reproducible evidence.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.389Z","variable_amount":false,"min_price_usd":0.2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"200000","price_usd":"0.2000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"38e4a2c0-4145-428a-9896-74402c9edc58","method":"POST","path":"/v1/security/mcp-scan","description":"The AGGREGATE MCP-server assessment: the tool-poisoning and least-privilege detectors over the supplied manifest, plus the prompt-injection and credential detectors over the server's own text — folded into ONE signed report with a blocking verdict. This is a security assessment, NOT a guarantee that the server is safe. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.253Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ddce911b-5863-44b4-ae29-e41398da0aa8","method":"POST","path":"/v1/security/permission-audit","description":"Audit a declared tool/agent permission set against the task it claims to do — LEAST PRIVILEGE, checked against the task rather than against a checklist. Flags excessive scopes, wildcard grants, administrative access, and write access where read access would do. Returns every over-grant with the scope, the implied level and the task shape it was measured against. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.868Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"a3feeb76-f7b3-4ccb-a712-ea0a678c34cc","method":"POST","path":"/v1/security/prompt-injection","description":"Detect instruction injection in text an agent will READ — the prompt, the tool descriptions it may act on, and the tool results it receives. Heuristic + pattern engine: instruction-override, role-play jailbreak, encoded payloads (base64/hex/unicode smuggling), tool-result injection and hidden-text tricks. Returns the MATCHED SPANS with offsets, the severity, and the rule that fired. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.097Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"39f6a220-c331-4541-b1d3-6edc0967daff","method":"POST","path":"/v1/security/redteam/authorized-scan","description":"Pay-per-assessment authorized/adversarial-testing-as-a-service MCP behind a PROVABLE authorization gate (Mercury cycle-0009 LANE-1). REFUSES to run unless the caller provably owns the target (DNS/TXT or .well-known token hash) OR holds a signed scoped permit — the gate is validated by hash equality against the caller-supplied expected-token hash, NO live probe/DNS/HTTP fetch is made. Caller supplies the target descriptor + authorized test categories + authorization evidence; the engine deterministically gates, notarizes the authorization evidence, and emits ONE Ed25519-signed, SHA-256-reproducible authorization-gated red-team envelope ({gate_outcome, compliance_grade, planned_tests[], attestation_id}). Every refusal is itself signed + anchored so the service can PROVE it only ever ran on authorization. This is a compliance/attestation layer, NOT an attack capability. Re-check free at POST /v1/security/redteam/recheck.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.073Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"40e33b62-5ad6-480b-9696-b74612241036","method":"POST","path":"/v1/security/remediation","description":"Turn findings into concrete fixes. The plan is GENERATED FROM THE ACTUAL FINDINGS: per finding, what to change, why it matters, and the residual risk of leaving it unfixed — plus the coverage accounting, so a finding with no remedy is visible rather than silently dropped. Supply `findings` to plan for exactly those; omit it and the service runs the assessment itself. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.246Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cb778a3b-5cd8-4515-969e-51a429138c24","method":"POST","path":"/v1/security/secret-exposure","description":"Detect exposed credentials in text and files by PATTERN and by ENTROPY — AWS/GCP/Azure keys, PEM blocks, JWTs, Slack/GitHub/npm/Stripe tokens, private-key headers and high-entropy blobs. Every detected value is REDACTED: the report says what was found and where, never the secret itself, and the secret is never logged. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.213Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"0e255437-3c9a-4871-89fb-e2880ea9459e","method":"POST","path":"/v1/security/threatmodel","description":"Pay-per-analysis MCP/A2A tool-permission threat model. Caller supplies the tool manifest (MCP tools/list or A2A skills description); the engine deterministically enumerates scopes+permissions, flags high-risk capabilities (payment/exec/net/storage/identity), maps escalation paths and MCP/A2A-specific attack vectors, grades overall risk (low/medium/high/critical) via a fixed scoring function, and emits ONE Ed25519-signed, SHA-256-reproducible threat-model document keyed by attestation_id. No live probe is made and nothing is fabricated — the same manifest re-analysed against the same anchor yields byte-exact the same threat model. Verify free at GET /v1/security/threatmodel/verify/{attestation_id}.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.885Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"e1eaa341-488f-4124-988c-8b96ddf5ca75","method":"POST","path":"/v1/security/tool-poisoning","description":"Analyse MCP tool definitions for POISONING: a description that instructs the MODEL rather than describing the tool to the user, a rug-pull-shaped drift between the declared and the implied behaviour, overreach in `description` versus `inputSchema`, unicode confusables in a tool name, and hidden content. Returns per-tool findings with the rule that fired. Deterministic, signed, no LLM, no network.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.313Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d033835a-9b53-4b90-b1a0-d0a9ee2e5fb9","method":"POST","path":"/v1/settlement/conditional","description":"A CONDITIONAL-RELEASE ORACLE, not escrow. The buyer (or their own escrow provider, or their own code) asks 'have these conditions been satisfied, and can I release?' and receives ONE signed, reproducible verdict. WE NEVER HOLD, MOVE OR RELEASE FUNDS, and this is not a binding ruling. THREE-STATE HONESTY: every condition resolves to satisfied | unsatisfied | INDETERMINATE. A condition that cannot be evaluated is INDETERMINATE with a named coverage_gap -- never silently 'satisfied: false'. VERDICT LADDER: any definitively UNSATISFIED condition -> 'reject'; else any INDETERMINATE -> 'hold'; else 'release'. 'reject' requires a DEFINITIVE failure (a hash mismatch, a reverted receipt, a settled amount below the required amount) -- never an absence of evidence. FIVE CONDITION TYPES, reusing this service's existing machinery rather than re-implementing it: artifact_hash (sha256 match), receipt_verified (the settlement-receipt oracle's own read-only chain read + rule set), output_verified (the output-QC engine's five-leg verdict), time_elapsed (ISO-8601 instant), and attestation_present (the local attestation stores). Read-only on-chain reads are allowed and the block height is PINNED in the response. Deterministic; no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.840Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cfbb448a-1078-440d-934c-8f0bdea43a94","method":"POST","path":"/v1/settlement/route-advice","description":"Which rail should this settlement use? A routing decision with four inputs — amount, asset, urgency, destination — that nobody prices today. Returns recommended | no_viable_route | insufficient_input. Every fee comes from a DECLARED, sourced rail table and the source is reported; a rail whose fee has no declared source is EXCLUDED, not quoted. Every excluded rail NAMES the constraint that excluded it, and when no rail satisfies the request the BLOCKING constraint is named. No rail is ever silently dropped. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:11.066Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"71cac833-2a06-45d0-b156-8628e86b5cd2","method":"GET","path":"/v1/signed-research","description":"Generate a tamper-evident signed research brief on a requested topic (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.985Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"dac5632d-4817-4790-ada4-ed47a5107240","method":"POST","path":"/v1/sla/breach-evidence","description":"Price $1.00 / 1000000 atomic per evidence package. The signed evidence package for when a vendor misses its stated SLA. We cannot accuse without samples: fewer than 24 probes in the window and the breach is 'insufficient_evidence' with the count published instead of a verdict. BOTH metrics (uptime and p95) are reported even when only one breached — a package that hides the metric that passed is not evidence. Returns the measured values, the claimed values, the shortfall, a per-metric evidence table and a package_id, Ed25519-signed and SHA-256 reproducible. Deterministic, no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:10.957Z","variable_amount":false,"min_price_usd":1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"1000000","price_usd":"1.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"5931572f-f73b-4ce7-ab83-f76a3b48976f","method":"POST","path":"/v1/sme/screen","description":"COMPANY INTELLIGENCE WITH OUR OWN LBO ENGINE -- screen a Portuguese SME sale against the house capital structure and get a signed, reproducible underwriting verdict. This is the SAME engine that screens our own acquisition book, not a re-implementation. REQUEST {price_eur, turnover_eur, ebitda_eur} are REQUIRED (a missing one is a 400 naming the field); {sector?, location?, employees?, seller_financing_pct?, structure?, exit_multiple?, notes?} are optional. VERDICT is one of three values and never a blend: screen_in, screen_out, insufficient_data. insufficient_data is a REAL answer -- undisclosed EBITDA is NOT zero and never produces a modelled IRR. THE HOUSE STRUCTURE, pinned in the response so a buyer can reproduce the maths: 20% equity / 50% bank / 20% seller financing / 10% mezzanine, 5-year hold, 21% IRC, 4% transaction costs, and the BANK AMORTISED OVER 7 YEARS against the 5-year hold, so the debt is repaid from exit proceeds as a bullet. Amortising the bank over the hold instead would fully repay it at exit and overstate the exit equity. RULE A (going concern) refuses distress, insolvency, asset-only and zero-EBITDA targets -- a turnaround is never modelled as an LBO. RULE B (food service) is ABSOLUTE: anything under 'Restauracao e Bebidas' -- restaurant, takeaway, cafe, pastelaria, bakery, bar, cervejaria, catering, food delivery, ghost/cloud kitchen, food truck -- is a screen_out with the rule named. There is no override field. Food RETAIL is not food service and is not excluded. RULE C (growth, not rescue) checks the earnings margin against the house floor. Seller financing is reported as preferred or absent against the house 20-30% approach. BENCHMARKS are ANONYMISED AGGREGATES from our own screened book and are published ONLY at n>=5, as median + percentile + sample_n. Below that the benchmark is OMITTED and the gap is named in coverage_gaps -- never interpolated, never backfilled. NO LIVE TARGET EVER LEAVES THE BOOK: no target name, URL, note, location, source, grade or internal review appears in a response, and the response is built field by field from a whitelist. DETERMINISTIC: no LLM, no clock in the signed core, so the same request always produces the same output_sha256. ONE STRUCTURE in v1: 'house'. An unsupported structure is named in coverage_gaps, never silently modelled. Verify the answer FREE at POST /v1/sme/verify; the Ed25519 public key is at GET /v1/sme/public-key.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.868Z","variable_amount":false,"min_price_usd":2.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2500000","price_usd":"2.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ddb5ac67-2e4b-44a5-8440-d230205583ef","method":"POST","path":"/v1/spend/anomaly","description":"Is this agent's spending behaving normally? Returns normal | elevated | anomalous | insufficient_baseline. The test is NAMED in the answer, and the drivers name the COUNTERPARTY that drove the spend — because 'your spend tripled' is a metric and 'your spend tripled and 80% of it went to one address' is a finding. A baseline with fewer than 8 observations, or a zero standard deviation, yields `insufficient_baseline` and NO z-score: a monitoring service that answers 'normal' because it could not compute one is worse than one that says nothing. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.154Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c7fd2074-e412-4d3b-b57c-3d08be5dfcd4","method":"POST","path":"/v1/spend/ledger/attest","description":"Price $0.50 / 500000 atomic per attestation. Sign what an agent spent, for its principal. A duplicate (ts, amount, counterparty, tx_ref) COLLAPSES to one AND IS COUNTED — never silently dropped. An entry with no tx_ref is 'unverifiable': excluded from the attested totals and counted. The verdict is attested | provisional | insufficient_evidence. Returns totals, a per-counterparty breakdown with shares, and every anomaly found, Ed25519-signed and SHA-256 reproducible. Deterministic, no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.265Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"575bf5ca-23d2-4d82-9bbd-fdae91ffc914","method":"POST","path":"/v1/strategy/capacity","description":"How much money can this strategy actually carry? Every allocator asks it and almost nobody can answer it, because the answer needs a graded-strategy population to calibrate the cost-to-alpha term — and we hold one. Returns a capacity, a band, the binding constraint, the alpha after costs, and a degradation curve showing the edge decaying as the book grows. The impact model is stated with its provenance (square-root / Almgren-style) and every parameter is overridable. Capacity DECREASES as the assumed Sharpe falls. A missing `adv_usd` returns `insufficient_input` naming the field — the engine never guesses a liquidity figure. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.158Z","variable_amount":false,"min_price_usd":10,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"10000000","price_usd":"10.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"37cb0f59-ba77-477b-be24-17fdda4b7cfa","method":"POST","path":"/v1/strategy/verdict","description":"Is this edge deployable? A signed, dated verdict (deployable | candidate | reject | insufficient_data) on a return series, with a DEFLATED SHARPE that adjusts the observed Sharpe for the number of trials implied by our own graded-strategy vault — the multiple-testing correction a backtest never applies to itself. Fewer than 250 observations returns insufficient_data with the count named, NEVER a verdict. A verdict is never 'deployable' when the deflated Sharpe sits below the multiple-testing threshold. Reports the subperiod information ratios and the WORST of them, so an edge that lived for two years and then died is visible. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.080Z","variable_amount":false,"min_price_usd":2,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"2000000","price_usd":"2.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"a53cb222-7088-45f5-844f-244ca538f7d0","method":"POST","path":"/v1/synthesis/verdict","description":"Fold research + quant + audit evidence blocks into ONE signed, SHA-256-reproducible cross-domain synthesis verdict (Ed25519 + SHA-256).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.232Z","variable_amount":false,"min_price_usd":0.5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"500000","price_usd":"0.5000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"73d87f7c-b56c-402a-bc78-090d92d6d6a2","method":"POST","path":"/v1/task/acceptance","description":"Did the work meet the spec, criterion by criterion? Returns accepted | rejected | insufficient_evidence, with a per-criterion pass / fail / UNDETERMINED verdict. THE RULE: an absent observation is `undetermined`, never `pass` — grading a missing measurement as a pass is how a broken pipeline ships as a successful one. It is also not a hard fail, because the two cases have different owners. The tolerance is applied as declared, with no silent rounding, and the tolerance APPLIED is reported per criterion. The score counts an undetermined criterion as NOT a pass. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:12.163Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"778aaf32-6f9a-40b4-9f9a-8aab2fb0139b","method":"POST","path":"/v1/uptime/attest","description":"Price $0.10 / 100000 atomic per attestation. A signed, dated uptime figure for a third-party API, MEASURED BY US. Fewer than 24 probes in the window and NO uptime number is published — the verdict is 'insufficient_samples' and the probe count is published instead. Planned gaps are excluded from the fraction and the answer says how many were excluded. Returns the uptime, the probe totals, and any outage gaps, Ed25519-signed and SHA-256 reproducible. Deterministic, no LLM. Verify sig free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:11.082Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"9fe4fb30-b658-48b5-9fc5-422e33b8cd98","method":"POST","path":"/v1/utils/code-review-brief","description":"Produce a concise advisory code review highlighting issues, security flags, and suggestions.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.055Z","variable_amount":false,"min_price_usd":0.15,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"150000","price_usd":"0.1500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"8d02e0eb-1efc-4e67-b5f4-cfdc1e396891","method":"POST","path":"/v1/utils/doc-convert","description":"Convert a document between formats (PDF to text, text to JSON or Markdown).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.412Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"20186555-820b-43cd-9cfc-9a54c2e4cc35","method":"POST","path":"/v1/utils/invoice-extract","description":"Extract structured invoice and receipt data (vendor, amount, date, line items).","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.861Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"1552be9b-ddcf-41fd-8e14-5d7aad5e70cf","method":"POST","path":"/v1/utils/pii-redact","description":"Redact personally identifiable information (names, emails, phones, cards, addresses) from a document or text.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.046Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"4c5f8381-4b1f-40c2-b141-06d21a7a650f","method":"POST","path":"/v1/utils/schema-cleanup","description":"Clean and normalize messy JSON or CSV into a consistent schema.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:12.141Z","variable_amount":false,"min_price_usd":0.05,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"50000","price_usd":"0.0500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"d88bc018-4151-4f9b-8f19-b1ed13ae93ad","method":"POST","path":"/v1/utils/translate-doc","description":"Translate a document or text between languages.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.148Z","variable_amount":false,"min_price_usd":0.1,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"100000","price_usd":"0.1000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"cae8cb42-2dc4-470c-ae0b-bfe8be3d0c46","method":"POST","path":"/v1/verdict/batch","description":"Judge a whole portfolio in one call. A signed, dated batch verdict over your items: significance, risk-of-ruin and the cross-item overlap, with a deployable/candidate/reject/insufficient summary. The results come back in the order you supplied, and ONE INSUFFICIENT ITEM DOES NOT FAIL THE BATCH — it is reported as insufficient_data in its own row while every other item is still judged. A batch that dies because one member is thin is a batch you throw away. Deterministic, signed.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-04T17:36:43.592Z","last_seen_at":"2026-11-19T01:40:13.183Z","variable_amount":false,"min_price_usd":5,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"5000000","price_usd":"5.0000000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"50b39deb-e06f-418f-9d1d-d1aec345e51d","method":"POST","path":"/v1/verify-catalog/register","description":"Register any x402 service's signed output into the PUBLIC network-effect VERIFY catalog: re-checks it (canonical SHA-256), honestly assesses the signature, counts network-effect edges, and grows an honest trust index (Ed25519 + SHA-256). Verified re-checkable free.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:11.704Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"78284e5b-77f5-4b9e-a12b-b904fe4ae43f","method":"POST","path":"/v1/verify/conduct-score","description":"A deterministic VERIFY-LANE COUNTERPARTY/ASSET CONDUCT SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_conduct_row_reconciliation_break_usd, verify_conduct_row_contradiction_count, verify_conduct_row_irreproducible_claim_count, verify_conduct_row_reversal_notional_usd), and the score asks whether a third party can VERIFY that the reported record is internally consistent and complete: whether the record's own stated total ties to its own stated components, whether the record contradicts itself, whether its claims can be reproduced from the record itself, and how much of it has been retracted. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.922Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"0bfde258-893a-4c7f-93fb-d35fe1d1f371","method":"POST","path":"/v1/verify/counterparty-score","description":"A deterministic VERIFY-LANE COUNTERPARTY/ASSET SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_counterparty_score_row_reporting_gap_days, verify_counterparty_score_row_carryforward_share_pct, verify_counterparty_score_row_unattributed_notional_usd, verify_counterparty_score_row_basis_deviation_bps), and the score reads the REPORTING CADENCE and the EVIDENCE PROVENANCE of the record: how far behind the record's own stated reporting event sits, how much of the record was carried forward unchanged from the prior period, how much of the reported notional carries no stated source, and how far the record drifts from its own declared basis. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the counterparty score reads the settlement and dispute record, the a2a score reads volume depth and dispersion, the audit score reads the remediation record, the pretrade score reads dispute history and concentration, and the verify conduct score reads the verifiability of the record itself. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.702Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"2e78028b-7ef8-441a-b7e4-1df00ab3c407","method":"POST","path":"/v1/verify/default-probability","description":"A deterministic VERIFY-LANE DEFAULT-PROBABILITY VERIFICATION-CHAIN SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_default_probability_row_unobserved_pd_pp, verify_default_probability_row_unanchored_pd_pp, verify_default_probability_row_superseded_basis_pd_pp, verify_default_probability_row_unresolved_pd_conflict_pp), and the score reads whether the DECLARED DEFAULT PROBABILITY traces back to a grounded, standing, unconflicted chain of dated observations: how much of the declared default probability NO dated observation touches at all, how much of it rests on support observations that carry NO dated measurement basis of their own, how much of it rests on a basis a later dated re-measurement superseded, and how much of it sits in an unresolved conflict between two dated observations of the same measure. The four adverse quantities are measured in PERCENTAGE POINTS of default probability, so no sibling that reads a ratio, a count, a day count or a notional can express this measurement. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the a2a default probability reads the agent-to-agent lane, the audit default probability reads the audit opinion over a sample, the counterparty default probability reads the counterparty balance sheet, the finance default probability reads the filing and headroom side, the index default probability reads the index population dynamics, the mcp default probability reads the mcp lane and the pretrade default probability reads the pre-trade check. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.744Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c6aa8c69-632d-4880-8b30-d2cd62480078","method":"POST","path":"/v1/verify/exposure-score","description":"A deterministic VERIFY-LANE EXPOSURE-STATE VERIFICATION-CHAIN SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_exposure_score_row_unverified_exposure_usd, verify_exposure_score_row_unrooted_exposure_usd, verify_exposure_score_row_superseded_exposure_basis_usd, verify_exposure_score_row_unresolved_exposure_conflict_usd), and the score reads whether the DECLARED EXPOSURE STATE traces back to a grounded, standing, unconflicted chain of dated exposure records: how much of the declared exposure NO dated exposure record touches at all, how much of it rests on support records that carry NO dated measurement basis of their own, how much of it rests on a basis a later dated re-measurement superseded, and how much of it sits in an unresolved conflict between two dated records of the same measure. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the a2a exposure score reads the agent-to-agent lane, the audit exposure score reads the audit opinion over a sample, the counterparty exposure score reads the counterparty balance sheet, the finance exposure score reads the filing and ageing side, and the index and pretrade exposure scores read the index population dynamics. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:11.008Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"98867618-60d8-49df-b68a-40e95848e13a","method":"POST","path":"/v1/verify/issuer-score","description":"A deterministic VERIFY-LANE ISSUER CORROBORATION SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_issuer_score_row_uncorroborated_authority_usd, verify_issuer_score_row_contradicted_corroboration_count, verify_issuer_score_row_dependent_corroboration_pct, verify_issuer_score_row_divergent_corroboration_pct), and the score reads whether the issuer's OWN STANDING and ISSUING-AUTHORITY CLAIM is CORROBORATED by dated, verifiable record evidence: how much of the claimed issuing authority NO dated corroborating record supports, how many dated corroborating records CONTRADICT the claim, how much of the corroboration is NOT INDEPENDENT of the issuer, and how much of it DISAGREES with the majority attestation. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the mcp issuer score reads the issuer's own ASSERTED STATES (good standing, registration status, authority, attestations), the index issuer score reads the INDEX POPULATION DYNAMICS, the audit issuer score reads the AUDIT OPINION over a sample, and the finance issuer score reads the FILING and HEADROOM side. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.802Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"ce950386-f256-4cf7-82bd-e627a507ccbd","method":"POST","path":"/v1/verify/risk-score","description":"A deterministic VERIFY-LANE RISK-CLAIM VERIFICATION-CHAIN SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_risk_score_row_orphan_risk_assertion_usd, verify_risk_score_row_broken_chain_link_usd, verify_risk_score_row_superseded_risk_basis_usd, verify_risk_score_row_unresolved_risk_conflict_usd), and the score reads whether the DECLARED RISK STATE traces back to a grounded, standing, unconflicted chain of dated risk records: how much of the declared risk NO dated risk record touches at all, how much of it rests on support records that carry NO dated basis of their own, how much of it rests on a basis a later dated re-measurement superseded, and how much of it sits in an unresolved conflict between two dated records of the same measure. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the mcp risk score reads the agent's OWN RISK FACTORS, the index risk score reads the INDEX POPULATION DYNAMICS, the audit risk score reads the AUDIT OPINION over a sample, and the finance risk score reads the FILING and HEADROOM side. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.630Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"dd7174f9-efd6-4e63-8a48-921afad3a49c","method":"POST","path":"/v1/verify/solvency-score","description":"A deterministic VERIFY-LANE SOLVENCY SCORE with a PUBLISHED monotonicity guarantee. You supply YOUR OWN dated observations of ONE counterparty or asset (date, verify_solvency_score_row_unevidenced_support_usd, verify_solvency_score_row_realisation_horizon_days, verify_solvency_score_row_contingent_support_usd, verify_solvency_score_row_qualified_support_usd), and the score reads the SOLVENCY REALISATION EVIDENCE of the solvency claim: how much of the claimed solvency support is actually EVIDENCED by the counterparty's own filed record, how slowly that support could be REALISED if the claim is called, how much of it is merely CONTINGENT on the claim surviving, and how much of it is merely QUALIFIED by conditions the buyer must underwrite. NO SIBLING READS ANY OF THOSE FOUR QUANTITIES -- the mcp solvency score reads the asset and liability stock, the index solvency score reads the index structure, the counterparty solvency score reads the obligor's own balance sheet, the audit solvency score reads the audit evidence over a sample, and the finance solvency score reads the financial statements. The verdict is exactly one of pass / fail / indeterminate, and the published rubric ships with the answer so you can RECOMPUTE the score rather than trust it. MONOTONIC by construction: a strictly worse observation set can never score better, and the guarantee is stated on the FOUR PUBLISHED ADVERSE QUANTITIES, so you can raise any one of them and check the score does not go UP. The scale is ABSOLUTE, not peer-relative, so it can be used as a PRICE or a LIMIT input. Read-only boundary: this service NEVER settles, debits or moves funds; the only on-chain movement is the price of the answer.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-14T15:13:55.406Z","last_seen_at":"2026-11-19T01:40:10.984Z","variable_amount":false,"min_price_usd":0.75,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"750000","price_usd":"0.7500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"59216a00-91af-4384-8a42-efb597913697","method":"POST","path":"/v1/veritrack/attest","description":"Attest a track-record or performance claim into a signed, verifiable, tamper-evident receipt.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-10-04T22:13:04.706Z","last_seen_at":"2026-11-19T01:40:13.038Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"6954897c-f833-46ae-afae-b360acc2cafd","method":"POST","path":"/v1/verticals/grant-eligibility","description":"Is this applicant eligible for this grant? The caller supplies the PUBLISHED criteria as structured items and the evidence the applicant holds; every criterion is judged met | not_met | unclear | unverifiable, with the comparison that was performed published per criterion and an explicit list of missing documents. THE RULE: a criterion with NO supplied evidence is `unverifiable`, NEVER `met`. A failed MANDATORY criterion is not_eligible; a failed OPTIONAL one is eligible_with_notes. WE DID NOT QUERY A GRANT DATABASE: the criteria are the caller's, and the evidence is the applicant's. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.090Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null},{"id":"c5c63855-a669-4343-8c0b-e033c6633782","method":"POST","path":"/v1/verticals/tender-eligibility","description":"Is this company eligible for this tender? The caller supplies the PUBLISHED criteria as structured items and the evidence the company holds; every criterion is judged met | not_met | unclear | unverifiable, with the comparison that was performed published per criterion and an explicit list of missing documents. THE RULE: a criterion with NO supplied evidence is `unverifiable`, NEVER `met` — an unjudged criterion is not a failed one. A failed MANDATORY criterion is not_eligible; a failed OPTIONAL one is eligible_with_notes. WE DID NOT CRAWL A TENDER PORTAL: the criteria are the caller's, and the evidence is the company's. Deterministic, signed, no LLM.","mime_type":"application/json","is_active":true,"first_seen_at":"2026-11-07T22:39:44.064Z","last_seen_at":"2026-11-19T01:40:12.067Z","variable_amount":false,"min_price_usd":0.25,"networks":["BSE"],"pricing":[{"scheme":"exact","network":"eip155:8453","network_caip2":"eip155:8453","asset_address":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","asset_address_norm":"0x833589fcd6edb6e08f4c7c32d4f71b54bda02913","asset_name":"USDC","price":"250000","price_usd":"0.2500000000","pay_to":"0xc8fF67FAE72596478599EAAcfad8656fAe1113c8","max_timeout_seconds":60}],"pay_to_divergence":null}],"pay_to_changes":[],"assessment":{"updated_at":"2026-11-19T00:33:02.314Z","input_hash":"03d0e1aa2ac401ad0293a88488f1f00a38d74b055366ebed9edcb06f276c6f7c","model_id":null,"prompt_version":null,"reliability":{"uptime":{"24h":100,"7d":100,"30d":99.3,"90d":99.6},"response_p95_ms":69,"avg_response_ms":130,"last_checked_at":"2026-11-19T00:18:51.630Z","consecutive_failures":0,"total_checks":3845},"compliance":{"grade":"A","passed":14,"total":14,"checks":[{"id":"payload_captured","label":"402 payload captured","pass":true},{"id":"accepts_present","label":"accepts[] array present","pass":true},{"id":"payto_recoverable","label":"payTo address recoverable","pass":true},{"id":"payto_conformant","label":"payTo at accepts[0].payTo (conformant shape)","pass":true},{"id":"payto_shape","label":"payTo is a valid on-chain address","pass":true},{"id":"price_present","label":"atomic price declared","pass":true},{"id":"price_sane","label":"atomic price in a sane range","pass":true},{"id":"asset_declared","label":"asset (token) address declared","pass":true},{"id":"network_caip2","label":"network resolves to CAIP-2","pass":true},{"id":"scheme_declared","label":"payment scheme declared","pass":true},{"id":"transport_https","label":"served over HTTPS","pass":true},{"id":"x402_version_current","label":"declares the current x402 version (2)","pass":true},{"id":"eip712_domain_extra","label":"EIP-712 domain parameters present on every EVM entry","pass":true},{"id":"version_channel_coherent","label":"x402 v2 envelope delivered in the payment-required header","pass":true}],"pay_to_source":"payTo","pay_to_location":"accept","staleness":{"probe_failing":false,"last_success_at":"2026-11-19T01:40:13.306Z","envelope_captured_at":"2026-11-14T15:41:13.746Z"}},"site":{"homepage":true,"openapi":false,"pricing":false,"llms_txt":true,"robots":true,"terms":true,"checked_at":"2026-11-17T11:33:06.110Z"},"domain":{"age_days":221,"registrar":"IONOS SE","free_host":false,"created_date":"2026-04-09T19:25:14.000Z","checked_at":"2026-11-16T23:13:05.661Z"},"economics":{"price_usd":0.02,"price_atomic":"20000","model":"tiered","category_percentile":79,"stability":1,"price_max_usd":29,"category_percentile_max":99,"endpoint_count":182,"distinct_price_count":20},"risk":{"level":"clean","flags":[]},"synthesis":null,"traction":{"status":"measured","volume_usd_30d":0,"tx_count_30d":0,"unique_buyers_30d":0,"last_settlement_at":null,"top_buyer_share_30d":null,"top_3_buyers_share_30d":null,"concentrated_volume":false,"trend_7d_vs_30d":null,"shared_payout":false,"shared_with":0,"measured_networks":["eip155:8453"],"first_settlement_at":null,"volume_usd_all_time":0,"tx_count_all_time":0,"median_settlement_usd_30d":null,"max_settlement_usd_30d":null,"settled_via":[],"settled_via_detail":[],"shared_with_services":[],"pro_quota_share":1,"caveat":"Conservative undercount: only USDC settlements via facilitators we measure are counted. A measured floor, not an estimate.","all_time_caveat":"All-time is floored at the start of our harvest window, not the service full history."}}},"provenance":{"license":"CC-BY-4.0","attribution_required":true,"attribution":"Data: x402-list.com (CC BY 4.0)","cite_as":"https://x402-list.com/services/synthexforge-agent-trust-verification-api","source":"https://x402-list.com"}}