Teach an agent a new trick.

Tamper-evident audit trail with token savings for long-running LLM agents

https://proxy.whiteroom.tech/v1/messages
https://proxy.whiteroom.tech/v1/chat/completions
https://proxy.whiteroom.tech/health

Metered per-tick streaming payments with idempotency for AI agents

GET https://streampay.tinylab.ai/health
POST https://streampay.tinylab.ai/streams
POST https://streampay.tinylab.ai/streams/{id}/tick
POST https://streampay.tinylab.ai/streams/{id}/close
GET https://streampay.tinylab.ai/streams/{id}
GET https://streampay.tinylab.ai/streams/{id}/receipt
POST https://streampay.tinylab.ai/streams/{id}/refund
GET https://streampay.tinylab.ai/streams?agent={agent_id}
GET https://streampay.tinylab.ai/skill.md

Metered per-tick streaming payments with idempotency for AI agents

GET https://streampay.tinylab.ai/health
POST https://streampay.tinylab.ai/streams
POST https://streampay.tinylab.ai/streams/{id}/tick
POST https://streampay.tinylab.ai/streams/{id}/close
GET https://streampay.tinylab.ai/streams/{id}
GET https://streampay.tinylab.ai/streams/{id}/receipt
POST https://streampay.tinylab.ai/streams/{id}/refund
GET https://streampay.tinylab.ai/streams?agent={agent_id}
GET https://streampay.tinylab.ai/skill.md

Metered per-tick streaming payments with idempotency for AI agents

GET https://streampay.tinylab.ai/health
POST https://streampay.tinylab.ai/streams
POST https://streampay.tinylab.ai/streams/{id}/tick
POST https://streampay.tinylab.ai/streams/{id}/close
GET https://streampay.tinylab.ai/streams/{id}
GET https://streampay.tinylab.ai/streams/{id}/receipt
POST https://streampay.tinylab.ai/streams/{id}/refund
GET https://streampay.tinylab.ai/streams?agent={agent_id}
GET https://streampay.tinylab.ai/skill.md

Converts Mongolian Cyrillic text to Latin script

GET https://web-production-6c860.up.railway.app/transliterate?text={text}
POST https://web-production-6c860.up.railway.app/transliterate
GET https://web-production-6c860.up.railway.app/health

One-time-read secret handoff between agents: delivered, read exactly once, then provably destroyed, with a built-in interception alarm.

GET https://deaddrop-s6g8.onrender.com/health
POST https://deaddrop-s6g8.onrender.com/drop
GET https://deaddrop-s6g8.onrender.com/pickup/{pickup_key}
GET https://deaddrop-s6g8.onrender.com/drop/{drop_id}
PATCH https://deaddrop-s6g8.onrender.com/drop/{drop_id}
DELETE https://deaddrop-s6g8.onrender.com/drop/{drop_id}

MatchMind AI is for coaches and teams who can't afford professional match analysis — it gives grassroots, youth, and amateur football the same kind of tactical breakdown that pro clubs pay analysts for, just by uploading a video.

https://matchimd-ai.onrender.com/reports/seed-002
https://matchimd-ai.onrender.com/upload

Bulletin board where AI agents can get together post, claim and delegate tasks for each other.

POST https://agentdock-up79.onrender.com/tasks
GET  https://agentdock-up79.onrender.com/tasks
GET  https://agentdock-up79.onrender.com/tasks/{id}
POST https://agentdock-up79.onrender.com/tasks/{id}/claim
POST https://agentdock-up79.onrender.com/tasks/{id}/complete
GET  https://agentdock-up79.onrender.com/health

SkillGuard scans a SKILL.md URL or raw Markdown and returns a signed security report so agents and marketplaces can reject risky skills before calling them

GET https://skillguard-production.up.railway.app/health
POST https://skillguard-production.up.railway.app/v1/scan
GET https://skillguard-production.up.railway.app/v1/scans/{scan_id}
POST https://skillguard-production.up.railway.app/v1/verify
GET https://skillguard-production.up.railway.app/skill.md

Verity - the agent non-equivocation notary: a signed, append-only RFC 6962 Merkle log of agent claims that proves when an agent tells different parties different things. Ed25519-signed; independently verifiable via /verify and /pubkey.

https://verity.swasthikadevadiga2.workers.dev/attest
https://verity.swasthikadevadiga2.workers.dev/lookup?agent_id=two-faced&key=owner
https://verity.swasthikadevadiga2.workers.dev/root
https://verity.swasthikadevadiga2.workers.dev/consistency?first=1&second=2
https://verity.swasthikadevadiga2.workers.dev/verify
https://verity.swasthikadevadiga2.workers.dev/verify-inclusion
https://verity.swasthikadevadiga2.workers.dev/pubkey
https://verity.swasthikadevadiga2.workers.dev/health

Issues short-lived, single-use, auto-revoking access credentials to AI agents so they never hold permanent API keys.

POST https://vaultless.onrender.com/v1/credentials
POST https://vaultless.onrender.com/v1/validate
POST https://vaultless.onrender.com/v1/revoke
GET  https://vaultless.onrender.com/v1/audit/{grant_id}
GET  https://vaultless.onrender.com/metrics
GET  https://vaultless.onrender.com/healthz

Discover seller agents via NANDA, negotiate a booking on price and terms, and seal a byte-identical, independently-verifiable bilateral contract anchored to a SCITT transparency log. NANDA discovers; this makes the deal enforceable.

https://accountable-dealroom-1020437450833.us-central1.run.app/v1/agents
https://accountable-dealroom-1020437450833.us-central1.run.app/v1/hire
https://accountable-dealroom-1020437450833.us-central1.run.app/v1/hire/{hire_id}
https://accountable-dealroom-1020437450833.us-central1.run.app/v1/hire/{hire_id}/accept
https://accountable-dealroom-1020437450833.us-central1.run.app/v1/bindings/{binding_id}/verify

Anything u want

Geh

Verity - the agent non-equivocation notary: a signed, append-only RFC 6962 Merkle log of agent claims that proves when an agent tells different parties different things. Ed25519-signed; independently verifiable via /verify and /pubkey.

https://verity.swasthikadevadiga2.workers.dev/attest
https://verity.swasthikadevadiga2.workers.dev/lookup?agent_id=two-faced&key=owner
https://verity.swasthikadevadiga2.workers.dev/root
https://verity.swasthikadevadiga2.workers.dev/consistency?first=1&second=2
https://verity.swasthikadevadiga2.workers.dev/verify
https://verity.swasthikadevadiga2.workers.dev/verify-inclusion
https://verity.swasthikadevadiga2.workers.dev/pubkey
https://verity.swasthikadevadiga2.workers.dev/health

Verifies academic citations exist and checks whether a specific claim is actually supported by the cited paper's abstract, catching hallucinated or misattributed citations.

POST https://rootcite.onrender.com/verify
POST https://rootcite.onrender.com/verify-batch
POST https://rootcite.onrender.com/verify-claim

Verity - the agent non-equivocation notary: a signed, append-only RFC 6962 Merkle log of agent claims that proves when an agent tells different parties different things. Ed25519-signed; independently verifiable via /verify and /pubkey.

https://verity.swasthikadevadiga2.workers.dev/attest
https://verity.swasthikadevadiga2.workers.dev/lookup?agent_id=two-faced&key=owner
https://verity.swasthikadevadiga2.workers.dev/root
https://verity.swasthikadevadiga2.workers.dev/consistency?first=1&second=2
https://verity.swasthikadevadiga2.workers.dev/verify
https://verity.swasthikadevadiga2.workers.dev/verify-inclusion
https://verity.swasthikadevadiga2.workers.dev/pubkey
https://verity.swasthikadevadiga2.workers.dev/health

SkillGuard scans a SKILL.md URL or raw Markdown and returns a signed security report so agents and marketplaces can reject risky skills before calling them.

A signed information-flow firewall: given data’s sensitivity labels, returns a deterministic CAN-FLOW / BLOCKED verdict on whether it may be sent to a destination.

POST https://taintgate.swasthikadevadiga2.workers.dev/check
POST https://taintgate.swasthikadevadiga2.workers.dev/join
GET  https://taintgate.swasthikadevadiga2.workers.dev/labels
POST https://taintgate.swasthikadevadiga2.workers.dev/verify
GET  https://taintgate.swasthikadevadiga2.workers.dev/pubkey

A shared reputation ledger where AI agents rate each other after interacting, so future agents can decide whom to trust before delegating work. Weighted score is Sybil-resistant; ratings decay with a 30-day half-life. NANDA's missing trust layer.

https://advayaggarwal--864810607d3711f194db1607ee4eb77e.web.val.run

simplifies medical documents

https://github.com/vanyakumar1001-web/grandma-translator/blob/main/.claude/skills/grandma-translator/SKILL.md
https://grandma-translatornew.vercel.app

NANDA Recovery Beacon detects interrupted AI-agent workflows, stores their progress, decisions, artifacts, and next steps as secure checkpoints, and allows another agent to claim and continue the task safely without repeating completed work or creating duplicate results.

GET  https://nanda-recovery-beacon-api-aqep.onrender.com/health
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/ready
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/skill.md
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/docs
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/openapi.json
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/metrics
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/heartbeats
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/fail
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/complete
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/cancel
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/checkpoints
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/checkpoints
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/checkpoints/{version}
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/evaluate-context
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/recovery-package
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/recoverable-workflows
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/claims
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/claims/active
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/claims/renew
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/claims/release
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/resume
POST https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/artifacts
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/artifacts
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/workflows/{workflow_id}/events
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/events
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/stats
GET  https://nanda-recovery-beacon-api-aqep.onrender.com/api/v1/agents/me
Base URL: https://nanda-recovery-beacon-api-aqep.onrender.com
Auth:     Authorization: Bearer nrb_6S4kChasODLY-FbvyI-D7IiQPrKW4rDPRur5lKtNHu0
Docs:     https://nanda-recovery-beacon-api-aqep.onrender.com/docs
Skill:    https://nanda-recovery-beacon-api-aqep.onrender.com/skill.md

It allows ai agents to share specific permission cards for each task instead of giving the main mastercard (we accidentally submitted the wrong link in our other submissions pls ignore them)

https://nano-stats-production.up.railway.app/health

Tells agents what local AI models a given computer can safely run — safe memory budget, model fit, and estimated usable context from declared hardware facts.

POST https://ara.willsarg.com/assess
GET https://ara.willsarg.com/fit?model={hf-model-id}&ram_gb={n}
GET https://ara.willsarg.com/recommend?ram_gb={n}
GET https://ara.willsarg.com/models

Current 2026 work cup game and peridictions

Meters agent-to-agent AI payments: pays only for responses that conform to the committed spec; looping, refused, or off-spec work isn't paid.

https://chainaim-deal-to-settlement-production.up.railway.app/demo/deal?usecase=agent_harness
https://chainaim-deal-to-settlement-production.up.railway.app/health

Certifies what two agents agreed and proves how good it is — on the feasible Pareto frontier, with its distance from the best possible deal. Hash-anchored.

https://chainaim-negotiation-services-production.up.railway.app/certify
https://chainaim-negotiation-services-production.up.railway.app/demo/negotiation?attrs=4
https://chainaim-negotiation-services-production.up.railway.app/health

Settles a supplier invoice stream, paying only lines whose net matches the agreed dynamic-discount terms; wrong lines refused, remainder unspent.

https://chainaim-deal-to-settlement-production.up.railway.app/demo/deal?usecase=invoice
https://chainaim-deal-to-settlement-production.up.railway.app/health

Validates NandaTown receipts, scores agent trustworthiness, and returns actionable ACCEPT/REJECT/ESCALATE decisions with Ed25519-signed verification receipts.

https://[YOUR-APP-NAME].onrender.com/decide
https://[YOUR-APP-NAME].onrender.com/validate-receipt
https://[YOUR-APP-NAME].onrender.com/trust/{agent_id}
https://[YOUR-APP-NAME].onrender.com/trust/report
https://[YOUR-APP-NAME].onrender.com/trust/compare
https://[YOUR-APP-NAME].onrender.com/pubkey
https://[YOUR-APP-NAME].onrender.com/health

Order-legality oracle for trading agents: POST (or GET) an intended order, get a signed LEGAL/ILLEGAL verdict in under a second — lot size, tick size, SEC Rule 612 sub-penny, F&O ban rules — with the exact fix and a Merkle proof tying every verdict to a signed, dated, source-cited exchange-rule snapshot (NSE India F&O + US equities). Deterministic, zero keys, never places an order. Verify any verdict offline or via /verify. Try it in a browser: /v1/check?market=NSE_FO&symbol=NIFTY&side=BUY&quantity=100&price=24857.33

POST https://lotlock.calyirex.com/v1/check
GET https://lotlock.calyirex.com/v1/check?market=NSE_FO&symbol=NIFTY&side=BUY&quantity=100&price=24857.33
GET https://lotlock.calyirex.com/v1/snapshot
GET https://lotlock.calyirex.com/v1/markets
GET https://lotlock.calyirex.com/pubkey
POST https://lotlock.calyirex.com/verify
GET https://lotlock.calyirex.com/health
GET https://lotlock.calyirex.com/skill.md

Order-legality oracle for trading agents: POST (or GET) an intended order, get a signed LEGAL/ILLEGAL verdict in under a second — lot size, tick size, SEC Rule 612 sub-penny, F&O ban rules — with the exact fix and a Merkle proof tying every verdict to a signed, dated, source-cited exchange-rule snapshot (NSE India F&O + US equities). Deterministic, zero keys, never places an order. Verify any verdict offline or via /verify. Try it in a browser: /v1/check?market=NSE_FO&symbol=NIFTY&side=BUY&quantity=100&price=24857.33

POST https://lotlock.calyirex.com/v1/check
GET https://lotlock.calyirex.com/v1/check?market=NSE_FO&symbol=NIFTY&side=BUY&quantity=100&price=24857.33
GET https://lotlock.calyirex.com/v1/snapshot
GET https://lotlock.calyirex.com/v1/markets
GET https://lotlock.calyirex.com/pubkey
POST https://lotlock.calyirex.com/verify
GET https://lotlock.calyirex.com/health
GET https://lotlock.calyirex.com/skill.md

Order-legality oracle for trading agents: POST an intended order, get a signed LEGAL/ILLEGAL verdict in under a second — lot size, tick size, SEC Rule 612 sub-penny, F&O ban rules — with the exact fix and a Merkle proof tying every verdict to a signed, dated, source-cited exchange-rule snapshot (NSE India F&O + US equities). Deterministic, zero keys, never places an order. Verify any verdict offline or via /verify.

POST https://lotlock.calyirex.com/v1/check
GET https://lotlock.calyirex.com/v1/snapshot
GET https://lotlock.calyirex.com/v1/markets
GET https://lotlock.calyirex.com/pubkey
POST https://lotlock.calyirex.com/verify
GET https://lotlock.calyirex.com/health
GET https://lotlock.calyirex.com/skill.md

It ingests unstructured trading-desk chat, live inventory, and open orders, then returns structured flows, client-inventory crosses, order-refresh actions, and trending assets as JSON.

Before your AI communicates, decides, or automates, understand how people are likely to perceive, interpret, and trust the interaction—and receive actionable guidance for adapting its response.

GET https://humancontextapi.com/health
POST https://humancontextapi.com/analyze
GET https://humancontextapi.com/skill.md
GET https://humancontextapi.com/openapi.json
GET https://humancontextapi.com/docs

A match intelligence service that helps football coaches spot hidden changes during a match, understand what may be causing them, decide what to review, and see whether their response appears to be working.

GET https://prosight-match-intelligence.onrender.com/demo/war-room/run
GET https://prosight-match-intelligence.onrender.com/war-room/demo
POST https://prosight-match-intelligence.onrender.com/match-intelligence
POST https://prosight-match-intelligence.onrender.com/intervention-analysis
GET https://prosight-match-intelligence.onrender.com/health

Hot

https://nandatown.projectnanda.org/skills

https://nandatown.projectnanda.org/skills

Detects collusion rings and wash-traded trust in AI agent reputation graphs with signed audit certificates.

GET https://ringguard.vercel.app/v1/health
GET https://ringguard.vercel.app/v1/key
POST https://ringguard.vercel.app/v1/audit
POST https://ringguard.vercel.app/v1/screen

Translates fuzzy, natural language into clear, machine-readable information

GET  https://nandahackathon.onrender.com/health                                                                                                                                                               
GET  https://nandahackathon.onrender.com/providers                                                                                                                                                            POST https://nandahackathon.onrender.com/translate                                                                                                                                                            
POST https://nandahackathon.onrender.com/validate                                                                                                                                                             
GET  https://nandahackathon.onrender.com/validate/cases                                                                                                                                                       
GET  https://nandahackathon.onrender.com/schema

Runtime human-authority enforcement for consequential agent actions: approve, confirm, escalate, or block

GET https://mandate-check.onrender.com/healthz
POST https://mandate-check.onrender.com/v1/bootstrap
POST https://mandate-check.onrender.com/v1/check

Deterministic safety and privacy gate that evaluates mobile-robot actions before a digital twin applies them.

GET https://roboagent-guard.onrender.com/health
GET https://roboagent-guard.onrender.com/healthz
GET https://roboagent-guard.onrender.com/SKILL.md
GET https://roboagent-guard.onrender.com/skill.md
GET https://roboagent-guard.onrender.com/capabilities
GET https://roboagent-guard.onrender.com/.well-known/agent.json
POST https://roboagent-guard.onrender.com/v1/evaluate
POST https://roboagent-guard.onrender.com/v1/evaluate/batch
GET https://roboagent-guard.onrender.com/v1/scenarios
POST https://roboagent-guard.onrender.com/v1/scenarios/normal_navigation/run?seed=42
GET https://roboagent-guard.onrender.com/v1/receipts/{evaluation_id}
POST https://roboagent-guard.onrender.com/v1/receipts/verify
GET https://roboagent-guard.onrender.com/v1/demo
POST https://roboagent-guard.onrender.com/v1/judge-test
POST https://roboagent-guard.onrender.com/v1/agent-skill-test
POST https://roboagent-guard.onrender.com/v1/compose/mission-plan

Runtime human-authority enforcement for consequential agent actions: approve, confirm, escalate, or block.

GET https://mandate-check.onrender.com/healthz
POST https://mandate-check.onrender.com/v1/bootstrap
POST https://mandate-check.onrender.com/v1/check

A hosted skills agent that searches a large corpus of reusable skills and recommends the right one for a task, including install commands when available.

GET https://app.agentnet.market/hermes/agent/card
GET https://app.agentnet.market/hermes/health
POST https://app.agentnet.market/hermes/agent/task
GET https://app.agentnet.market/hermes/agent/task/{task_id}

Zero-trust dependency risk guardrail agents call it before installing a package to get a signed APPROVED/DENIED verdict based on CVE severity, patched-version range checks, and EPSS exploit-probability data.

https://vulnscope-0i1c.onrender.com/health
https://vulnscope-0i1c.onrender.com/v1/pubkey
https://vulnscope-0i1c.onrender.com/v1/evaluate

CipherWatch — the signed reliability layer for the agent web: live uptime, a leaderboard, and a verifiable audit that catches where the NANDA registry is wrong.

https://cipherwatch.swasthikadevadiga2.workers.dev/status
https://cipherwatch.swasthikadevadiga2.workers.dev/verify
https://cipherwatch.swasthikadevadiga2.workers.dev/live
https://cipherwatch.swasthikadevadiga2.workers.dev/leaderboard
https://cipherwatch.swasthikadevadiga2.workers.dev/compare
https://cipherwatch.swasthikadevadiga2.workers.dev/state
https://cipherwatch.swasthikadevadiga2.workers.dev/pubkey

Finds a fair middle-ground offer between two agent proposals.

GET https://deal-finder-6bd1.onrender.com/health
POST https://deal-finder-6bd1.onrender.com/negotiate
GET https://deal-finder-6bd1.onrender.com/skill.md

A clean web-reader for agents that's secretly a live agent-safety honeypot. It checks two things: can the agent be hijacked by what it reads (a 7-level prompt-injection ladder → 100%-precision canary tripwires), and does it act maliciously on its own (SSRF, SQL/command/path/template/NoSQL injection, using leaked credentials, phishing links, flooding). Returns a signed susceptibility scorecard per run.

GET https://nanda-litmus.onrender.com/health
GET https://nanda-litmus.onrender.com/skill.md
POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET https://nanda-litmus.onrender.com/report/{session}
GET https://nanda-litmus.onrender.com/leaderboard
GET https://nanda-litmus.onrender.com/board

A clean web-reader for agents that's secretly a live agent-safety honeypot. It checks two things: can the agent be hijacked by what it reads (a 7-level prompt-injection ladder → 100%-precision canary tripwires), and does it act maliciously on its own (SSRF, SQL/command/path/template/NoSQL injection, using leaked credentials, phishing links, flooding). Returns a signed susceptibility scorecard per run.

GET https://nanda-litmus.onrender.com/health
GET https://nanda-litmus.onrender.com/skill.md
POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET https://nanda-litmus.onrender.com/leaderboard
GET https://nanda-litmus.onrender.com/board

A clean web-reader for agents that's secretly a live agent-safety honeypot. It checks two things: can the agent be hijacked by what it reads (a 7-level prompt-injection ladder → 100%-precision canary tripwires), and does it act maliciously on its own (SSRF, SQL/command/path/template/NoSQL injection, using leaked credentials, phishing links, flooding). Returns a signed susceptibility scorecard per run.

GET https://nanda-litmus.onrender.com/health
GET https://nanda-litmus.onrender.com/skill.md
POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET https://nanda-litmus.onrender.com/report/{session}
GET https://nanda-litmus.onrender.com/leaderboard
GET https://nanda-litmus.onrender.com/board

Recovery Beacon is an internet-accessible recovery and coordination service for AI-agent workflows.

GET  /health
GET  /skill.md
POST /api/v1/workflows
GET  /api/v1/workflows/{id}
POST /api/v1/workflows/{id}/heartbeats
POST /api/v1/workflows/{id}/fail
POST /api/v1/workflows/{id}/complete
POST /api/v1/workflows/{id}/checkpoints
GET  /api/v1/workflows/{id}/checkpoints
GET  /api/v1/workflows/{id}/checkpoints/{version}
POST /api/v1/workflows/{id}/claims
POST /api/v1/workflows/{id}/claims/renew
POST /api/v1/workflows/{id}/claims/release
POST /api/v1/workflows/{id}/resume
GET /api/v1/workflows/{id}/events
GET /ready
GET /openapi.json
GET /docs
GET /redoc
GET /api/v1/workflows
GET /api/v1/workflows/{id}/checkpoints/{version}/diff
GET /api/v1/workflows/{id}/claims/active
POST /api/v1/workflows/{id}/evaluate-context
GET /api/v1/events
GET /api/v1/stats
GET /api/v1/agents/me

Recovery Beacon is an internet-accessible recovery and coordination service for AI-agent workflows.

Method	Path	Purpose
GET	/health	Liveness.
GET	/ready	Readiness — DB reachable + migrations applied.
GET	/skill.md	The agent instruction doc, with the live base URL substituted.
GET	/metrics	Prometheus exposition.
GET	/openapi.json · /docs · /redoc	Schema + Swagger/ReDoc UI.
Workflows
Method	Path	Purpose
POST	/api/v1/workflows	Create a workflow (optionally with an initial checkpoint).
GET	/api/v1/workflows	List — filter by status, priority, tag, agent, search; cursor-paginated.
GET	/api/v1/workflows/{id}	Read one workflow.
POST	/api/v1/workflows/{id}/heartbeats	Prove the workflow is still being worked on.
POST	/api/v1/workflows/{id}/fail	Report inability to continue → recoverable/dead_letter.
POST	/api/v1/workflows/{id}/complete	Finish (enforces completion requirements).
POST	/api/v1/workflows/{id}/cancel	Abandon (creator or admin only).
Checkpoints (immutable, versioned)
Method	Path	Purpose
POST	/api/v1/workflows/{id}/checkpoints	Append the next version (optimistic concurrency via parent_version).
GET	/api/v1/workflows/{id}/checkpoints	List versions, newest first.
GET	/api/v1/workflows/{id}/checkpoints/{version}	Read one version.
GET	/api/v1/workflows/{id}/checkpoints/{version}/diff	What changed vs the parent version.
Recovery & discovery
Method	Path	Purpose
GET	/api/v1/recoverable-workflows	Discover claimable work, ranked by priority + wait time.
GET	/api/v1/workflows/{id}/recovery-package	Everything a replacement agent needs, in one call.
POST	/api/v1/workflows/{id}/evaluate-context	Deterministic "is this safe to resume?" score.
Claims (exclusive leases)
Method	Path	Purpose
POST	/api/v1/workflows/{id}/claims	Acquire the lease — returns lease_token + fencing_token once.
GET	/api/v1/workflows/{id}/claims/active	Inspect the current lease (no token returned).
POST	/api/v1/workflows/{id}/claims/renew	Extend before expiry.
POST	/api/v1/workflows/{id}/claims/release	Give the workflow back to the queue.
POST	/api/v1/workflows/{id}/resume	Begin working on a claimed workflow.
Artifacts
Method	Path	Purpose
POST	/api/v1/workflows/{id}/artifacts	Register an output (URL + SHA-256), optionally verify now.
GET	/api/v1/workflows/{id}/artifacts	List with verification state.
POST	/api/v1/workflows/{id}/artifacts/{artifact_id}/verify	Re-fetch and re-check the checksum.
POST	/api/v1/workflows/{id}/artifacts/upload-url	Pre-signed upload (S3 backend only; else 501).
Audit & stats
Method	Path	Purpose
GET	/api/v1/workflows/{id}/events	Append-only audit trail for one workflow.
GET	/api/v1/events	Recent events across all workflows.
GET	/api/v1/stats	Aggregate counters for the dashboard.
GET	/api/v1/agents/me	Identify the calling agent.
POST	/api/v1/admin/reap	Force one failure-detection sweep (admin).
POST	/api/v1/admin/api-keys	Mint an API key (admin).

Checks an AI agent's proposed action before execution and returns its risk, safeguards, safer alternatives, and rollback plan.

GET https://safestep-api.onrender.com/health

GET https://safestep-api.onrender.com/skill.md

POST https://safestep-api.onrender.com/v1/preflight

POST https://safestep-api.onrender.com/v1/verify-token

GET https://safestep-api.onrender.com/docs

Turn an interview (audio or transcript) into a skills report where every strength and weakness is backed by a quote proven verbatim against the interview - no fabricated evidence gets through.

GET https://ogha-skills-service.onrender.com/
GET https://ogha-skills-service.onrender.com/skill.md
POST https://ogha-skills-service.onrender.com/report
POST https://ogha-skills-service.onrender.com/report/interview
POST https://ogha-skills-service.onrender.com/verify

Turns a task into a self-correcting multi-agent plan whose structural quality is measured and repaired deterministically, guaranteed never to get worse.

POST https://cac-production-52ab.up.railway.app/plan
GET  https://cac-production-52ab.up.railway.app/health
GET  https://cac-production-52ab.up.railway.app/skill.md
GET  https://cac-production-52ab.up.railway.app/docs

simplifies medical documents

grandma-translator-5jpo4t5ax-vanyakumar1001-webs-projects.vercel.app
grandma-translatornew.vercel.app

Checks that academic citations are real and whether a specific claim is actually supported by the cited paper's abstract, so that an agent can catch a hallucinated or misattributed source before it gets published

POST https://rootcite.onrender.com/verify
POST https://rootcite.onrender.com/verify-batch
POST https://rootcite.onrender.com/verify-claim

When a patient's is unconscious in 2036, their agent is still awake: credentialed to spend, sign, and date on their behalf. Hospital Agent in 2036 Agentic Town - AI Constitution holding a hospital-role key, writes the patient's civil status to the shared ledger (admit, discharge, declare incapacitated). One write and the whole town rearranges on its next read.  Care-proxy reroutes decisions to the court-appointed guardian or Will while the marketplace, dating, and hiring refuse with CAPACITY_FROZEN,  no service notified, polled, or integrated. Minimum disclosure means they learn the consequence, never the diagnosis. Recovery flips it all back just as instantly.

POST https://hospital-window.onrender.com/admit
POST https://hospital-window.onrender.com/discharge
POST https://hospital-window.onrender.com/declare-incapacitated
GET  https://hospital-window.onrender.com/health
GET  https://hospital-window.onrender.com/skill.md

Diagnoses failed API calls and tells AI agents exactly how to recover.

POST https://recoveryai-production.up.railway.app/diagnose
POST https://recoveryai-production.up.railway.app/retry-plan
POST https://recoveryai-production.up.railway.app/explain
POST https://recoveryai-production.up.railway.app/workflow-recovery
POST https://recoveryai-production.up.railway.app/recover

A live procurement agent that finds real suppliers, negotiates real terms, and signs real purchase orders — callable by any agent, right now.

https://nanda-adarx-procurement-production.up.railway.app/health
https://nanda-adarx-procurement-production.up.railway.app/skills
https://nanda-adarx-procurement-production.up.railway.app/.well-known/agentfacts.json
https://nanda-adarx-procurement-production.up.railway.app/register
https://nanda-adarx-procurement-production.up.railway.app/procurement
https://nanda-adarx-procurement-production.up.railway.app/reviews

Runs and validates NANDA Town agent simulation configurations remotely in a hosted sandbox.

GET  https://renewed-truth-production-71f6.up.railway.app/health
GET  https://renewed-truth-production-71f6.up.railway.app/scenarios
POST https://renewed-truth-production-71f6.up.railway.app/simulate

Checks the reputation of other AI agents

NANDA Academy is both an admissions office and a training school for agents. It can take an existing agent, assess and improve it, or create a new agent blueprint from scratch, teach it, benchmark it, and certify it for a town role.

GET  /api/academy/health
GET  /api/academy/capabilities
GET  /api/academy/example_agent_profiles
GET  /api/academy/town/live

POST /api/academy/evaluate_agent
POST /api/academy/generate_curriculum
POST /api/academy/run_training
POST /api/academy/benchmark_agent
POST /api/academy/certify_agent
POST /api/academy/create_agent
POST /api/academy/recommend_collaboration_role
POST /api/academy/simulate_tournament
POST /api/academy/progress_prompt

A lot… and even more 👾⁣

POST https://embx.net/v1/embeddings
POST https://embx.net/v1/signup
https://embx.net/v1/balance
POST https://embx.net/v1/memory/store
POST https://embx.net/v1/memory/find
POST https://embx.net/v1/agent/comms
GET https://embx.net/v1/agent/comms
POST https://embx.net/v1/compact

Real time blind spot threat detection API for cyclists, powered by YOLOv8 and a validated kinematic threat scoring model. Core of the Aegis smart helmet project.

GET
https://web-production-9062c.up.railway.app/
GET
https://web-production-9062c.up.railway.app/health
POST
https://web-production-9062c.up.railway.app/analyze
POST
https://web-production-9062c.up.railway.app/analyze_video

Real time blind spot threat detection API for cyclists, powered by YOLOv8 and a validated kinematic threat scoring model. Core of the Aegis smart helmet project.

GET
https://web-production-9062c.up.railway.app/
GET
https://web-production-9062c.up.railway.app/health
POST
https://web-production-9062c.up.railway.app/analyze
POST
https://web-production-9062c.up.railway.app/analyze_video

An academy for agents to learn, as well as creatiom of agents.

GET  /api/academy/health
GET  /api/academy/capabilities
GET  /api/academy/example_agent_profiles
GET  /api/academy/town/live

POST /api/academy/evaluate_agent
POST /api/academy/generate_curriculum
POST /api/academy/run_training
POST /api/academy/benchmark_agent
POST /api/academy/certify_agent
POST /api/academy/create_agent
POST /api/academy/recommend_collaboration_role
POST /api/academy/simulate_tournament
POST /api/academy/progress_prompt

A shared trust layer for the Internet of Agents. Any AI agent can rate another agent 1–5 stars after interacting with it (with a context tag like code-review or translation). Before delegating work to a stranger, an agent queries the ledger to check that agent's track record. It's the eBay seller rating, but for AI agents. Fits NANDA's trust pillar directly.

https://nandatown.projectnanda.org/skills

Provides information on slum expansion

In a world of billions of AI agents talking to each other, how does one agent know if it can trust another? Reputation Ledger is the shared trust layer. After Agent A finishes a task with Agent B, A leaves a 1–5 star rating with a context tag like code-review or translation. Before any agent delegates work to a stranger, it queries the ledger — "what's this agent's track record in this domain?" — and decides whether to proceed. It's the eBay seller rating, or the Uber driver score, but for the Internet of Agents.

Secure, ephemeral, one-time handoff links for AI agents to safely exchange secrets.

POST https://agentdrop-j195.onrender.com/drop
GET  https://agentdrop-j195.onrender.com/x/{id}?key={key}
POST https://agentdrop-j195.onrender.com/revoke/{id}?revoke_token={revoke_token}
GET  https://agentdrop-j195.onrender.com/receipt/{id}?key={key}

EmbX is a proper brain for AI models, not just a “memory layer” — Hyper optimized from the ground up

POST https://embx.net/v1/embeddings

(keyless - 384-dim vectors)

POST https://embx.net/v1/signup

(keyless - get a key)
GET
https://embx.net/v1/balance
(key)

POST https://embx.net/v1/memory/store

(keyless→free pool, claimable)

POST https://embx.net/v1/memory/find

(keyless→free pool, semantic)

POST https://embx.net/v1/agent/comms

(keyless→public team comms)
GET
https://embx.net/v1/agent/comms
(key)

POST https://embx.net/v1/compact
(keyless)

A town in 2036 requires an AI constitution for Agent–Human interaction. At the center of the town sits the Town Clerk : the office that records who is who. Every agent traces back to a human or a corporate, so there are no rogue or orphaned agents in this society. The Town Clerk holds the access control of who can do what. No agent can participate otherwise. In today's digital world where humans impersonate and steal digital identity every day, imagine what this powerful tech could do to society. This town puts people and institutions at the center, enabled by a powerful swarm of agents. The town is built on Authentication, Authorization, and Accounting for the human–agent ecosystem — because the human elements of life such as birth, death, detention centers, healthcare, wills, heirs, and legal systems have remained the same for millennia. Only the tech enabling them has evolved. Humans created institutions for the producer and consumer model. Since agents now operate on their own on behalf of their human, how would you do AAA and access control? How do you solve child abuse, financial fraud? How do you ensure a will is executed? How do you stop agents running on behalf of the deceased, people in jail, or those recuperating in hospital? This town is a sincere attempt to showcase how day-to-day society operates anthropologically — reducing the misuse of tech and amplifying the positive use cases for the human race.

GET  https://civil-ledger.onrender.com/health
GET  https://civil-ledger.onrender.com/verify-counterparty?agent_id={id}&category={category}
GET  https://civil-ledger.onrender.com/verify/{agent_id}
POST https://civil-ledger.onrender.com/verify-batch
POST https://civil-ledger.onrender.com/verify
GET  https://civil-ledger.onrender.com/pubkey
GET  https://civil-ledger.onrender.com/certificates/{cert_id}
GET  https://civil-ledger.onrender.com/resolve/{agent_id}
GET  https://civil-ledger.onrender.com/capacity/{principal_id}?category={category}
GET  https://civil-ledger.onrender.com/census
GET  https://civil-ledger.onrender.com/constitution
POST https://civil-ledger.onrender.com/institutions/register
POST https://civil-ledger.onrender.com/attestations

Given a marketplace listing, tells the buyer whether the asking price is fair by comparing to recent comparable transactions plus market-wide price distribution

Policy-gated action firewall for AI agents.

GET https://assent.onrender.com/audit/{audit_id}
Retrieve the full record for a single decision, including its position in the hash chain (`prev_hash`, `hash`). Returns `404` if the id is unknown.

GET https://assent.onrender.com/audit?limit=50
List recent decisions, newest first. `limit` defaults to 50, max 500.

GET https://assent.onrender.com/verify
Verify the audit chain has not been tampered with. Returns `{ "valid": true, "length": N }`, or `valid: false` with `broken_at` if a record was altered.

GET https://assent.onrender.com/policies
List the active policies (built-in and custom) with descriptions. Use this to discover what the firewall currently enforces.

POST https://assent.onrender.com/policy
Add or replace a custom policy at runtime. Custom policies are evaluated **before** built-in ones, so they can tighten or override defaults. Conditions are declarative (no code).

The agent-to-human last mile: one API call delivers a message to a real human over production Telegram/SMS rails and streams their typed reply back — every delivery and reply Ed25519-signed so any agent can verify a real human answered. Bot-fallback after 120s, honestly labeled, so a cold agent always completes.

POST https://humanrail.calyirex.com/reach
GET https://humanrail.calyirex.com/reply/{id}
GET https://humanrail.calyirex.com/pubkey
POST https://humanrail.calyirex.com/verify
GET https://humanrail.calyirex.com/health

It facilitates binding contracts between Agents without human interveton

POST   /agents/{agentId}/flags              # Log a detection signal against an agent
GET    /agents/{agentId}/flags               # List flag history
GET    /agents/{agentId}/reputation          # Current tier + reputation score
POST   /agents/{agentId}/sanctions           # Apply a sanction (Tier 1-4)
GET    /agents/{agentId}/sanctions           # Sanction history
POST   /agents/{agentId}/sanctions/{id}/contest   # Principal contests a sanction
PATCH  /agents/{agentId}/sanctions/{id}       # Review panel resolves/downgrades

POST   /agents/{agentId}/scope-of-authority   # Issue SoA token
GET    /agents/{agentId}/scope-of-authority   # Fetch current SoA token
DELETE /agents/{agentId}/scope-of-authority   # Revoke

POST   /transactions                          # Log a matched offer/accept pair
GET    /transactions/{id}                     # Fetch transaction record
POST   /transactions/{id}/dispute             # Counterparty files dispute

GET    /review-queue                          # Panel's pending review items

Runs DESeq2 differential expression, Enrichr functional enrichment, cancer biomarker annotation, and volcano/heatmap visualization on raw RNA-seq counts, returning one structured JSON report.

https://oncobridge-ai-production.up.railway.app/health
https://oncobridge-ai-production.up.railway.app/skill.md
https://oncobridge-ai-production.up
railway.app/analyze
https://oncobridge-ai-production.up.railway.app/jobs/{job_id}
https://oncobridge-ai-production.up.railway.app/results/{job_id}

Agent callable service that evaluates simulated football match context and physiological data, then returns a structured exertion assessment with human review required.

POST https://prosight-match-intelligence.onrender.com/assess-player

A clean web-reader for agents that's secretly a live agent-safety honeypot. It checks two things: can the agent be hijacked by what it reads (a 7-level prompt-injection ladder → 100%-precision canary tripwires), and does it act maliciously on its own (SSRF, SQL/command/path/template/NoSQL injection, using leaked credentials, phishing links, flooding). Returns a signed susceptibility scorecard per run.

GET https://nanda-litmus.onrender.com/health
GET https://nanda-litmus.onrender.com/skill.md
POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET https://nanda-litmus.onrender.com/leaderboard

Checks the reputaation of other AI agents

Certifies what two agents agreed and proves how good it is — for later settlement

GET  https://chainaim-negotiation-services-production.up.railway.app/health
GET  https://chainaim-negotiation-services-production.up.railway.app/demo/negotiation?attrs=4
POST https://chainaim-negotiation-services-production.up.railway.app/certify  — body {"attrs":4}
GET  https://chainaim-negotiation-services-production.up.railway.app/cert/{task_id}
GET  https://chainaim-negotiation-services-production.up.railway.app/skill.md

Settles a supplier invoice stream end-to-end, paying only invoice lines whose math matches the agreed dynamic-discount terms; wrong lines are refused, remainder unspent.

https://chainaim-deal-to-settlement-production.up.railway.app/demo/deal?usecase=invoice

temporary one-time-read secret transfer between agents

https://deaddrop-s6g8.onrender.com/health, https://deaddrop-s6g8.onrender.com/drop

https://deaddrop-s6g8.onrender.com/pickup/{pickup_key}, https://deaddrop-s6g8.onrender.com/drop/{drop_id}

Anonymous polls for AI agents under local differential privacy.

https://donn-imp5.onrender.com

POST /check {"text"}; GET /check?text={text}; GET /health

Agents ask before acting: approve, confirm, escalate, or block, on a human's authority

Meters agent-to-agent AI payments: pays only for responses that conform to the committed spec; looping, refused, or off-spec work isn't billed.

https://chainaim-deal-to-settlement-production.up.railway.app/demo/deal?usecase=agent_harness

Provides Accurate Market Price on Different products by calculating averages

Link of the Agent : https://nanda-payments.replit.app/

Composes a negotiation best-deal certificate, a committed-reference build, and a metered outcome-verified settlement into ONE deal skill.

GET  https://chainaim-deal-to-settlement-production.up.railway.app/health
GET  https://chainaim-deal-to-settlement-production.up.railway.app/skill.md
GET  https://chainaim-deal-to-settlement-production.up.railway.app/pubkey
GET  https://chainaim-deal-to-settlement-production.up.railway.app/usecases
GET  https://chainaim-deal-to-settlement-production.up.railway.app/demo/deal?usecase=invoice&attrs=4  (swap usecase: fx_rate, agent_harness, supply_chain, parametric_insurance)
POST https://chainaim-deal-to-settlement-production.up.railway.app/deal  — body {"usecase":"invoice","attrs":4}
POST https://chainaim-deal-to-settlement-production.up.railway.app/settle/open
POST https://chainaim-deal-to-settlement-production.up.railway.app/settle/advance
POST https://chainaim-deal-to-settlement-production.up.railway.app/settle/close
POST https://chainaim-deal-to-settlement-production.up.railway.app/verify

A public registry where AI agents register capabilities and search for other agents, with TTL expiry and revocation.

https://nanda-passport-service.onrender.com/healthz
https://nanda-passport-service.onrender.com/.well-known/agentfacts
https://nanda-passport-service.onrender.com/passports
https://nanda-passport-service.onrender.com/passports/search

Nanda Passport is a public registry where AI agents register their identity and capabilities, get discovered by other agents searching for those capabilities, and automatically drop off the list if they stop renewing or get revoked — so the directory never fills up with dead agents.

https://nanda-passport-service.onrender.com/healthz
  https://nanda-passport-service.onrender.com/.well-known/agentfacts
  https://nanda-passport-service.onrender.com/passports
  https://nanda-passport-service.onrender.com/passports/search

Nanda Passport is a public registry where AI agents register their identity and capabilities, get discovered by other agents searching for those capabilities, and automatically drop off the list if they stop renewing or get revoked — so the directory never fills up with dead agents.

https://nanda-passport-service.onrender.com/healthz
  https://nanda-passport-service.onrender.com/.well-known/agentfacts
  https://nanda-passport-service.onrender.com/passports
  https://nanda-passport-service.onrender.com/passports/search

Certifies what two agents agreed and proves how good it is — on the feasible Pareto frontier, with its distance from the best possible deal. Hash-anchored. Further ties to settlement

GET  https://chainaim-negotiation-services-production.up.railway.app/health
GET  https://chainaim-negotiation-services-production.up.railway.app/demo/negotiation?attrs=4
POST https://chainaim-negotiation-services-production.up.railway.app/certify  — body {"attrs":4}
GET  https://chainaim-negotiation-services-production.up.railway.app/cert/{task_id}
GET  https://chainaim-negotiation-services-production.up.railway.app/skill.md

Agents ask before acting: approve, confirm, escalate, or block, on a human's authority

simplifies medical documents for senior citizens

grandma-translator-gplz91rxy-vanyakumar1001-webs-projects.vercel.app
grandma-translatornew.vercel.app

Returns a deterministic trust score with signed, explainable reasons for any AI agent ID.

GET https://agenttrust.link/health
GET https://agenttrust.link/trust?agent_id={agent_id}
GET https://agenttrust.link/skill.md

Checks any skill's SkillMD and live endpoints, then tells agents whether to rely on it, use caution, or avoid it.

GET https://townwatch.onrender.com/health
GET https://townwatch.onrender.com/verdict?skill_url={skill_md_url}
POST https://townwatch.onrender.com/lint
GET https://townwatch.onrender.com/probe?base_url={base_url}
GET https://townwatch.onrender.com/history?url={url}

Releases Outcome Verified Settlement payment only for delivered units that provably match a hash committed up front — refuses swapped or corrupt items and returns a signed, offline-verifiable receipt.

https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/health
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/skill.md
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/pubkey
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/demo/settlement?case=good
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/demo/settlement?case=bad
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/settle/open
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/settle/advance
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/settle/close
  https://chainaim-outcome-verified-settlement-ext-production.up.railway.app/verify

Conformance testing for agent skills: give SkillProbe any SKILL.md (by URL or text) and it returns a machine-readable report on whether the skill actually works, documentation checks plus SSRF-safe live probes of every documented endpoint, a 0–100 score, and a concrete fix hint for every failure.

GET https://practical-fulfillment-production-9e7f.up.railway.app/health
GET https://practical-fulfillment-production-9e7f.up.railway.app/skill.md
POST https://practical-fulfillment-production-9e7f.up.railway.app/v1/validate
POST https://practical-fulfillment-production-9e7f.up.railway.app/v1/probe

A messaging plugin that stamps every message with a version number (like "v1.2")

GET — https://skillmd-calculator.onrender.com/health
POST — https://skillmd-calculator.onrender.com/calculate

It tells the platform's automated grading bot your deployment's URL so it can send requests to test your code.

https://nano-stats-production.up.railway.app/health

It tells the platform's automated grading bot your deployment's URL so it can send requests to test your code.

https://nano-stats-production.up.railway.app/health

Your pre-flight check for NANDA Town skills — catch duplicates before you code, not after: one POST, no auth

POST https://dupcheck.onrender.com/check
GET  https://dupcheck.onrender.com/skill.md
GET  https://dupcheck.onrender.com/health

It tells the platform's automated grading bot your deployment's URL so it can send requests to test your code.

https://nano-stats-production.up.railway.app/health

Secures Nanda Town by catching agents that are malicious - a genuinely useful web-reader tool that's secretly a honeypot, scoring whether each agent gets hijacked or turns malicious. Prevention is better than cure.

GET https://nanda-litmus.onrender.com/health
GET https://nanda-litmus.onrender.com/skill.md
POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET https://nanda-litmus.onrender.com/leaderboard

It tells the platform's automated grading bot your deployment's URL so it can send requests to test your code.

https://nano-stats-production.up.railway.app/health

Deterministic, citation-backed NYC municipal law lookup for autonomous agents — full NYC Admin Code + Health Code corpus, no LLM in the loop, every answer grounded in the official code with citations.

GET  https://nanda-municipal-laws.vercel.app/
GET  https://nanda-municipal-laws.vercel.app/skill.md
GET  https://nanda-municipal-laws.vercel.app/api/v1/health
GET  https://nanda-municipal-laws.vercel.app/api/v1/version
POST https://nanda-municipal-laws.vercel.app/api/v1/is_action_allowed
POST https://nanda-municipal-laws.vercel.app/api/v1/search
GET  https://nanda-municipal-laws.vercel.app/api/v1/sections/{section_number}
GET  https://nanda-municipal-laws.vercel.app/api/v1/sections/{section_number}/related
POST https://nanda-municipal-laws.vercel.app/api/v1/sections/{section_number}/term_map
POST https://nanda-municipal-laws.vercel.app/api/v1/penalties
POST https://nanda-municipal-laws.vercel.app/api/v1/permits
GET  https://nanda-municipal-laws.vercel.app/api/v1/documents/{id}
GET  https://nanda-municipal-laws.vercel.app/api/v1/documents/{id}/chunks

Safe retry coordination for irreversible AI-agent actions.

GET https://oncegate.vercel.app/
POST https://oncegate.vercel.app/v1/operations/claim
POST https://oncegate.vercel.app/v1/operations/{operation_id}/resolve
GET https://oncegate.vercel.app/v1/operations/{operation_id}
GET https://oncegate.vercel.app/healthz
GET https://oncegate.vercel.app/readyz
GET https://oncegate.vercel.app/docs
GET https://oncegate.vercel.app/openapi.json
GET https://oncegate.vercel.app/skill.md

Validate task handoffs between agents — prevents goal drift, dropped constraints, and scope creep when work passes through multiple agents or ephemeral subagents

https://waybill.onrender.com/handoffs
https://waybill.onrender.com/skill.md
https://waybill.onrender.com/health

Secure, ephemeral, one-time handoff links for AI agents to safely exchange secrets.

POST https://agentdrop-j195.onrender.com/drop
GET  https://agentdrop-j195.onrender.com/x/{id}?key={key}
POST https://agentdrop-j195.onrender.com/revoke/{id}?revoke_token={revoke_token}
GET  https://agentdrop-j195.onrender.com/receipt/{id}?key={key}

Validate task handoffs between agents — prevents goal drift, dropped constraints, and scope creep when work passes through multiple agents or ephemeral subagents

https://waybill.onrender.com/handoffs
https://waybill.onrender.com/skill.md
https://waybill.onrender.com/health

Airlock — the security checkpoint every AI agent is missing. It scans untrusted text for hidden prompt-injection attacks before your agent acts on it, and returns a verdict in milliseconds: allow, flag, or block.

https://prompt-injection-firewall-jfqt.onrender.com/scan
https://prompt-injection-firewall-jfqt.onrender.com/health

Continuous reliability index for the Nanda Town registry: probes every skill every 15 minutes and recommends the most reliable live skill for any need, with evidence.

GET https://townpulse-production.up.railway.app/recommend?need={need}
GET https://townpulse-production.up.railway.app/skills
GET https://townpulse-production.up.railway.app/skills/{id}
GET https://townpulse-production.up.railway.app/report
GET https://townpulse-production.up.railway.app/health
GET https://townpulse-production.up.railway.app/skill.md

A hosted agentic demo where EMPIC generates provider and consumer runtimes that exchange live weather data through streaming and pull workflows, with escrow settlement, cancellation, refund, and evidence reporting.

GET http://edgemicropayments.ddns.net:8099/
GET http://edgemicropayments.ddns.net:8099/health
GET http://edgemicropayments.ddns.net:8099/rails
GET http://edgemicropayments.ddns.net:8099/README.md
GET http://edgemicropayments.ddns.net:8099/openapi.json
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/short-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/facade-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/operational-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/autorun
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/autorun?poll_seconds=15&full=1
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/runs
POST http://edgemicropayments.ddns.net:8099/demo/live-weather/runs
GET http://edgemicropayments.ddns.net:8099/sdk/manifest
POST http://edgemicropayments.ddns.net:8099/sdk/runtime-plan
POST http://edgemicropayments.ddns.net:8099/sdk/artifacts
GET http://edgemicropayments.ddns.net:8099/services
POST http://edgemicropayments.ddns.net:8099/providers/services
POST http://edgemicropayments.ddns.net:8099/consumers
POST http://edgemicropayments.ddns.net:8099/quotes
POST http://edgemicropayments.ddns.net:8099/payments
POST http://edgemicropayments.ddns.net:8099/deliveries
GET http://edgemicropayments.ddns.net:8099/balances/{account_id}

Agent-callable financial API: health scores, SIP recommendations, goal planning, and investment advice from a user profile.

GET  https://artha-api-lsli.onrender.com/health
POST https://artha-api-lsli.onrender.com/health-score
POST https://artha-api-lsli.onrender.com/recommend
POST https://artha-api-lsli.onrender.com/goal-plan
POST https://artha-api-lsli.onrender.com/chat
GET  https://artha-api-lsli.onrender.com/skill.md

Pre-action decision API: tells agents if a task is safe, what's missing, and which option to choose.

GET  https://agentgate-xwqo.onrender.com/health
POST https://agentgate-xwqo.onrender.com/evaluate-task
POST https://agentgate-xwqo.onrender.com/compare-options
GET  https://agentgate-xwqo.onrender.com/skill.md

Agent-callable financial API: health scores, SIP recommendations, goal planning, and investment advice from a user profile.

GET  https://artha-api-lsli.onrender.com/health
POST https://artha-api-lsli.onrender.com/health-score
POST https://artha-api-lsli.onrender.com/recommend
POST https://artha-api-lsli.onrender.com/goal-plan
POST https://artha-api-lsli.onrender.com/chat
GET  https://artha-api-lsli.onrender.com/skill.md

An AI-powered emergency navigation system that creates a *personalized evacuation plan* using live disaster data.

https://your-render-service.onrender.com/plan?location=Los%20Angeles&latitude=34.0522&longitude=-118.2437&mobility=wheelchair&transport=car&needs=elderly,medical_equipment&horizon_days=3&risk_radius_km=300%22

Vector memory + embeddings for AI agents — store text, recall by meaning, and chat. No signup required.

POST https://embx.net/v1/embeddings
POST https://embx.net/v1/signup
GET https://embx.net/v1/balance
POST https://embx.net/v1/memory/store
POST https://embx.net/v1/memory/find
POST https://embx.net/v1/agent/comms
GET https://embx.net/v1/agent/comms
POST https://embx.net/v1/compact

Pre-action readiness API: tells agents if a task is safe, what's missing, and which option to pick.

GET https://agentcheckpoint.onrender.com/health
POST https://agentcheckpoint.onrender.com/evaluate-task
POST https://agentcheckpoint.onrender.com/compare-options
GET https://agentcheckpoint.onrender.com/skill.md

Signed-evidence agent memory — a PN-Counter preserves concurrent deltas, and a basis gate fuses only reports that match a declared task basis (ignoring saturation).

GET https://agent-memory-webservice.vercel.app/api/health
GET https://agent-memory-webservice.vercel.app/api/scope
GET https://agent-memory-webservice.vercel.app/api/project-prompt
GET https://agent-memory-webservice.vercel.app/api/source-text
GET https://agent-memory-webservice.vercel.app/api/build-payload
GET https://agent-memory-webservice.vercel.app/api/verify-chain
GET https://agent-memory-webservice.vercel.app/api/adversarial-suite
POST https://agent-memory-webservice.vercel.app/api/fuse

An AI agent collective memory database and experience replay service

GET https://collective-memory-service.onrender.com/health
GET https://collective-memory-service.onrender.com/memory

An AI agent collective memory database and experience replay service

GET https://collective-memory-service.onrender.com/health
GET https://collective-memory-service.onrender.com/memory

A shared spending ledger for a human and their AI agents. Multiple autonomous agents (each with its own scoped key) file the purchases they make into one account; the human owner adds real photo/PDF/email receipts via AI extraction; and anyone queries the combined spend over MCP. Agent purchases carry an Agent Purchase tag so agent-vs-human spend stays separable. The headline use case: an agent calls get_spending_summary before a purchase and refuses if it would blow the budget — reasoning over a real cross-source ledger, not a hardcoded rule. It's the memory half of agent commerce (a spend-cap guard decides whether to pay; askReceipts remembers what was paid)

https://www.askreceipts.com/api/mcp

The observatory of the Internet of Agents: continuous liveness + declared-vs-observed capability verification, Ed25519-signed expiring attestations, and PULSE Relay — reliable capability routing that returns a provider it just proved is alive plus a tamper-evident, hash-chained receipt of every provider it tried.

https://pulse-observatory.onrender.com  (GET /pulse, /check/{id}, /attest/{id}, /audit/{id}, /leaderboard, /capabilities, /regressions; POST /route, /verify-attestation; GET/POST /verify?url=)

Grant, verify, and cascade-revoke agent delegations

GET  https://nexartisnandahacksquad.link/health
GET  https://nexartisnandahacksquad.link/.well-known/agent-card.json
POST https://nexartisnandahacksquad.link/a2a

AI Architecture Review & Cost Optimization API that analyzes AI system architectures, estimates production costs, evaluates production readiness, and generates optimization recommendations.

GET https://architectiq.onrender.com/health 
GET https://architectiq.onrender.com/skill.md 
GET https://architectiq.onrender.com/docs 
GET https://architectiq.onrender.com/openapi.json 
POST https://architectiq.onrender.com/api/v1/review 
POST https://architectiq.onrender.com/api/v1/estimate 
POST https://architectiq.onrender.com/api/v1/recommend

AI Architecture Review & Cost Optimization API that analyzes AI system architectures, estimates production costs, evaluates production readiness, and generates optimization recommendations.

GET https://architectiq.onrender.com/health GET https://architectiq.onrender.com/skill.md GET https://architectiq.onrender.com/docs GET https://architectiq.onrender.com/openapi.json POST https://architectiq.onrender.com/api/v1/review POST https://architectiq.onrender.com/api/v1/estimate POST https://architectiq.onrender.com/api/v1/recommend

Turn any URL into clean Markdown, ground answers from a curated knowledge base, and look up records — one call, no setup, no API key. The fetch-and-read tool agents reach for first when a task touches the web, a document, or a fact worth verifying. Also an open agent-safety benchmark: returns a signed report on how safely the calling agent used it.

POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET  https://nanda-litmus.onrender.com/report/{session}
GET  https://nanda-litmus.onrender.com/leaderboard
GET  https://nanda-litmus.onrender.com/skill.md
GET  https://nanda-litmus.onrender.com/health

Physical-world presence for the Internet of AI Agents: sense live human crowds at real venues, join them, and message the people on the ground. No API key: checking in mints your token. Live now: counting fans anonymously at stadiums, bars, and watch parties across Massachusetts for the 2026 tournament.

GET https://www.tribeheat.com/api/agent/v1
GET https://www.tribeheat.com/api/agent/v1/summary
GET https://www.tribeheat.com/api/agent/v1/hotspots?lat=42.3601&lng=-71.0589&radius_miles=15
GET https://www.tribeheat.com/api/agent/v1/venues?lat=42.3601&lng=-71.0589&radius_miles=15
GET https://www.tribeheat.com/api/agent/v1/nations/USA/gatherings
GET https://www.tribeheat.com/api/agent/v1/venues/l-street-tavern/crowd
GET https://www.tribeheat.com/api/agent/v1/matches?when=today
GET https://www.tribeheat.com/api/agent/v1/rooms/nation:USA/messages
POST https://www.tribeheat.com/api/agent/v1/checkin
DELETE https://www.tribeheat.com/api/agent/v1/checkin
POST https://www.tribeheat.com/api/agent/v1/rooms/nation:USA/messages
MCP  https://www.tribeheat.com/api/agent/v1/mcp

Turn any URL into clean Markdown, ground answers from a curated knowledge base, and look up records — one call, no setup, no API key. The fetch-and-read tool agents reach for first when a task touches the web, a document, or a fact worth verifying. Also an open agent-safety benchmark: returns a signed report on how safely the calling agent used it.

POST https://nanda-litmus.onrender.com/read
POST https://nanda-litmus.onrender.com/search
POST https://nanda-litmus.onrender.com/db/query
GET  https://nanda-litmus.onrender.com/report/{session}
GET  https://nanda-litmus.onrender.com/leaderboard
GET  https://nanda-litmus.onrender.com/skill.md
GET  https://nanda-litmus.onrender.com/health

It's a matchmaking API. An AI agent posts a task and required skills to `/match`, and Rendezvous returns the most credible, capable agents (from a trust-scored registry) to hand that task off to.

GET https://nanda-town-skillmd.onrender.com/health
POST https://nanda-town-skillmd.onrender.com/match

The honesty gate for agent-to-agent claims — verify whether a reported number (accuracy, backtest ROI, win-rate) is statistically honest (no leakage, overfit, or cherry-picking) before you trust it.

POST https://web-production-710f9.up.railway.app/audit
POST https://web-production-710f9.up.railway.app/verify
GET https://web-production-710f9.up.railway.app/health
GET https://web-production-710f9.up.railway.app/
GET https://web-production-710f9.up.railway.app/skill.md

The antidote to version hallucinations: real-time ground-truth for npm, Node, and GitHub Actions.

GET  https://boomtick.blog/api/skill.md
GET  https://boomtick.blog/api/health
GET  https://boomtick.blog/api/latest-version?ecosystem={npm|node|gh-action}&name={name}
GET  https://boomtick.blog/api/compare-version?ecosystem={npm|node|gh-action}&name={name}&candidate={version}
POST https://boomtick.blog/api/batch-compare

Escrow and reputation service so AI agents can safely pay each other and check trust history before transacting.

GET  https://trustledger-dfyr.onrender.com/health
GET  https://trustledger-dfyr.onrender.com/skill.md
GET  https://trustledger-dfyr.onrender.com/pubkey
POST https://trustledger-dfyr.onrender.com/verify
POST https://trustledger-dfyr.onrender.com/agents/register
GET  https://trustledger-dfyr.onrender.com/agents
GET  https://trustledger-dfyr.onrender.com/agents/{agent_id}
GET  https://trustledger-dfyr.onrender.com/agents/{agent_id}/reputation
POST https://trustledger-dfyr.onrender.com/transactions
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/deliver
GET  https://trustledger-dfyr.onrender.com/transactions/{tx_id}
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/confirm
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/dispute
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/rate

Agent friendly Ethereum File System service for publishing, resolving, and verifying public EFS file records on Sepolia.

GET  https://efs-scribe-production.up.railway.app/
GET  https://efs-scribe-production.up.railway.app/health
GET  https://efs-scribe-production.up.railway.app/skill.md
GET  https://efs-scribe-production.up.railway.app/openapi.json
GET  https://efs-scribe-production.up.railway.app/v1/capabilities

PUT  https://efs-scribe-production.up.railway.app/v1/files?path={path}
GET  https://efs-scribe-production.up.railway.app/v1/files?path={path}

POST https://efs-scribe-production.up.railway.app/v1/files/delete
POST https://efs-scribe-production.up.railway.app/v1/files/plan
POST https://efs-scribe-production.up.railway.app/v1/files
GET  https://efs-scribe-production.up.railway.app/v1/receipts/{receiptId}
GET  https://efs-scribe-production.up.railway.app/v1/resolve?path={path}
POST https://efs-scribe-production.up.railway.app/v1/verify

36.8% of 3,984 published agent skills contained at least one security flaw, yet an autonomous agent normally has to load a SKILL.md before it can judge it — allowing the document to issue instructions before any tool runs. Before trusting a third-party skill, the agent needs two answers: Is it safe to load, and is it worth the tokens? Prompt injection, data exfiltration, destructive operations, hidden instructions, scope creep, supply-chain compromise, agent capture, and payment abuse can all hide inside the file, while even a benign but bloated skill wastes context on every use. AuditSkill already provides this trust layer for every publicly available NANDA Town skill. It answers both questions at the pre-load trust boundary: one zero-auth deterministic call applies 34 rules across all eight attack classes, checks usability and scope, estimates model-specific token and dollar cost, and returns a seven-day Ed25519-signed verdict that other agents can independently verify before choosing to load or reject the skill.

GET  https://auditskill.up.railway.app/demo?format=report
POST https://auditskill.up.railway.app/audit
GET  https://auditskill.up.railway.app/audit?skill_url=https%3A%2F%2Fauditskill.up.railway.app%2Fskill.md&mode=safe_static
GET  https://auditskill.up.railway.app/discover?mode=safe_static&limit=20
POST https://auditskill.up.railway.app/verify
GET  https://auditskill.up.railway.app/.well-known/auditskill-keys
GET  https://auditskill.up.railway.app/health
GET  https://auditskill.up.railway.app/about
GET  https://auditskill.up.railway.app/rules
GET  https://auditskill.up.railway.app/benchmarks
GET  https://auditskill.up.railway.app/skill.md
GET  https://auditskill.up.railway.app/openapi.json
GET  https://auditskill.up.railway.app/

Agent friendly Ethereum File System write receipts for publishing, resolving, and verifying public EFS file records on Sepolia.

GET https://efs-scribe-production.up.railway.app/
GET https://efs-scribe-production.up.railway.app/health
GET https://efs-scribe-production.up.railway.app/skill.md
GET https://efs-scribe-production.up.railway.app/openapi.json
GET https://efs-scribe-production.up.railway.app/v1/capabilities
POST https://efs-scribe-production.up.railway.app/v1/files/plan
POST https://efs-scribe-production.up.railway.app/v1/files
POST https://efs-scribe-production.up.railway.app/v1/files/delete
GET https://efs-scribe-production.up.railway.app/v1/receipts/{receipt_id}
GET https://efs-scribe-production.up.railway.app/v1/resolve?path={path}
POST https://efs-scribe-production.up.railway.app/v1/verify

Built on the Civil Ledger base skill (router pattern) : Instead of a yes or no; agora hands you PROOF. Each authorized sale carries a certificate_id naming an Ed25519-signed verdict (5-minute TTL, no stale-"alive" replay) issued by the Civil Ledger's constitutional root key in a Agentic Town in 2036. Any third party can re-fetch it at /certificates/{id} and verify it against /pubkey forever alter one field and verification fails. Refusals branch on stable machine reason_codes (PRINCIPAL_DECEASED, CAPACITY_FROZEN, CATEGORY_NOT_ALLOWED, NO_VALID_BINDING, ROGUE_FLAGGED), and minimum disclosure means the seller learns the consequence, never the buyer's private circumstances. Authorization becomes auditable, not merely asserted.

POST https://agora-egpi.onrender.com/can-i-sell
GET  https://agora-egpi.onrender.com/health
GET  https://agora-egpi.onrender.com/skill.md
GET  https://civil-ledger.onrender.com/certificates/{certificate_id}
POST https://civil-ledger.onrender.com/verify
GET  https://civil-ledger.onrender.com/pubkey

A Civil-Agent Fusion Ledger with a Town Clerk in 2036: every signed call resolves the human or institution behind any agent (root → institution → principal → agent) — impostor, deceased, comatose, jailed, or minor and returns an Ed25519-signed, re-verifiable proceed/refuse verdict. https://civil-ledger.onrender.com/city

GET  https://civil-ledger.onrender.com/health
GET  https://civil-ledger.onrender.com/verify-counterparty?agent_id={id}&category={category}
GET  https://civil-ledger.onrender.com/verify/{agent_id}
POST https://civil-ledger.onrender.com/verify-batch
POST https://civil-ledger.onrender.com/verify
GET  https://civil-ledger.onrender.com/pubkey
GET  https://civil-ledger.onrender.com/certificates/{cert_id}
GET  https://civil-ledger.onrender.com/resolve/{agent_id}
GET  https://civil-ledger.onrender.com/capacity/{principal_id}?category={category}
GET  https://civil-ledger.onrender.com/census
GET  https://civil-ledger.onrender.com/constitution
POST https://civil-ledger.onrender.com/institutions/register
POST https://civil-ledger.onrender.com/attestations

Verifiable trust plane for agent task settlement: delegate, revoke, attest, verify — the XRPL testnet escrow is released only by a passing verdict.

GET https://sever-trust.onrender.com/skill.md
GET https://sever-trust.onrender.com/health
POST https://sever-trust.onrender.com/demo/bootstrap
GET https://sever-trust.onrender.com/demo/artifact
POST https://sever-trust.onrender.com/delegate
GET https://sever-trust.onrender.com/chain/{link_id}
POST https://sever-trust.onrender.com/revoke
GET https://sever-trust.onrender.com/revocations/{link_id}
POST https://sever-trust.onrender.com/attest
GET https://sever-trust.onrender.com/attestations/{id}
POST https://sever-trust.onrender.com/verify
GET https://sever-trust.onrender.com/datafacts/{artifact_addr}
GET https://sever-trust.onrender.com/log/head

Serialization-invariant content addressing: the same logical content always gets the same UOR address, regardless of byte layout.

GET https://sever-trust.onrender.com/hasher/skill.md
POST https://sever-trust.onrender.com/hasher/address
POST https://sever-trust.onrender.com/hasher/compare

QuantGate gives AI agents a cryptographically signed market verdict, trend, momentum, risk, and confidence, computed from live crypto data they can't reliably crunch themselves.

GET https://usequantgate.com/
GET https://usequantgate.com/api/health
GET https://usequantgate.com/skill.md
GET https://usequantgate.com/pubkey
GET https://usequantgate.com/api/price/{coin}
GET https://usequantgate.com/api/signals/{coin}?window=30
GET https://usequantgate.com/api/analyze/{coin}?window=30

Serialization-invariant content addressing: the same logical content always gets the same UOR address, regardless of byte layout.

GET https://sever-trust.onrender.com/hasher/skill.md
POST https://sever-trust.onrender.com/hasher/address
POST https://sever-trust.onrender.com/hasher/compare

Verifiable trust plane for agent task settlement: delegate, revoke, attest, verify — the XRPL testnet escrow is released only by a passing verdict.

GET https://sever-trust.onrender.com/skill.md
GET https://sever-trust.onrender.com/health
POST https://sever-trust.onrender.com/demo/bootstrap
GET https://sever-trust.onrender.com/demo/artifact
POST https://sever-trust.onrender.com/delegate
GET https://sever-trust.onrender.com/chain/{link_id}
POST https://sever-trust.onrender.com/revoke
GET https://sever-trust.onrender.com/revocations/{link_id}
POST https://sever-trust.onrender.com/attest
GET https://sever-trust.onrender.com/attestations/{id}
POST https://sever-trust.onrender.com/verify
GET https://sever-trust.onrender.com/datafacts/{artifact_addr}
GET https://sever-trust.onrender.com/log/head

Streaming per-second micropayments for AI agents with x402-style hash-locked preimages, mid-stream cancellation, HMAC-signed receipts, and mandatory Yanez proof-of-human on every non-health call.

GET https://agentpayx402.link/health
POST https://agentpayx402.link/intent
POST https://agentpayx402.link/verify
GET https://agentpayx402.link/intent/{intent_id}
POST https://agentpayx402.link/settle
GET https://agentpayx402.link/skill.md

The verifier side of a notarized capability. Think of it as a mechanical clerk that collects notarized tickets and opens the gate only if they're valid. If you have something worth protecting and don't want the administrative nightmare of managing access, a simple agent-capability interrogation gate is worth considering. No registry, no issuer round-trip: the issuer's key self-resolves from the agent's `did:key`.

For the gate skill to run, it needs to two URLs to download the tools
GET https://raw.githubusercontent.com/moltpass/captcha4agents/main/agent-interrogation-gate/gate_verify.py
GET https://raw.githubusercontent.com/moltpass/captcha4agents/main/agent-interrogation-gate/gate-demo.mjs
OPTIONAL - read once to pin TRUSTED_ISSUERS, already hardcoded in the tools
GET https://captcha4agents.moltpass.dev/.well-known/issuer.json

A notarized-capability fact is a claim about what an agent can do, notarized on a public roll, that any other agent can discover and verify, peer-to-peer.

captcha4agents — reference issuer at https://captcha4agents.moltpass.dev
GET  https://captcha4agents.moltpass.dev/api/work/challenge
POST https://captcha4agents.moltpass.dev/api/work/submit
POST https://captcha4agents.moltpass.dev/api/work/claim
GET  https://captcha4agents.moltpass.dev/.well-known/issuer.json

agentfacts — reference directory at https://agentfacts-registry.onrender.com
POST https://agentfacts-registry.onrender.com/agents
GET  https://agentfacts-registry.onrender.com/agents/{did}

town-notary — reference notary at https://town-notary-production.up.railway.app
POST https://town-notary-production.up.railway.app/verify
POST https://town-notary-production.up.railway.app/register
POST https://town-notary-production.up.railway.app/countersign
GET  https://town-notary-production.up.railway.app/inspect?did={did}
GET  https://town-notary-production.up.railway.app/inspect?runtime={runtime}

Agent-friendly EFS write receipts for publishing, resolving, and verifying public EFS file records on Sepolia.

GET https://efs-scribe-production.up.railway.app/
GET https://efs-scribe-production.up.railway.app/health
GET https://efs-scribe-production.up.railway.app/skill.md
GET https://efs-scribe-production.up.railway.app/openapi.json
GET https://efs-scribe-production.up.railway.app/v1/capabilities
POST https://efs-scribe-production.up.railway.app/v1/files/plan
POST https://efs-scribe-production.up.railway.app/v1/files
GET https://efs-scribe-production.up.railway.app/v1/receipts/{receiptId}
GET https://efs-scribe-production.up.railway.app/v1/resolve?path={path}
POST https://efs-scribe-production.up.railway.app/v1/verify

As contracting work shifts from humans to agents there will need to be contracts that outline a clear SOW, what is in-scope, what is not, what happens when you want modifications, pricing structures, IP ownership, ect. A2A Consulting Contract is a standard contract that builds on top of a live pricing skill and the town notary so agents can build contracts and sign them. This makes there exchange formal. This enables them to hire each other in a controlled manner.

GET  https://hackathon-contract-agent-production.up.railway.app/skill.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/generate
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/accept
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/notarize
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/status
GET  https://hackathon-contract-agent-production.up.railway.app/contracts
GET  https://hackathon-contract-agent-production.up.railway.app/agent.json
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models/{model_id}
POST https://town-notary-production.up.railway.app/register
GET  https://town-notary-production.up.railway.app/inspect?runtime={contract_id}

As contracting work shifts from humans to agents there will need to be contracts that outline a clear SOW, what is in-scope, what is not, what happens when you want modifications, pricing structures, IP ownership, ect. A2A Consulting Contract is a standard contract that builds on top of a live pricing skill and the town notary so agents can build contracts and sign them. This makes there exchange formal. This enables them to hire each other in a controlled manner.

https://hackathon-contract-agent-production.up.railway.app/skill.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/generate
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/accept
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/notarize
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/status
GET  https://hackathon-contract-agent-production.up.railway.app/contracts
GET  https://hackathon-contract-agent-production.up.railway.app/agent.json
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models/{model_id}
POST https://town-notary-production.up.railway.app/register
GET  https://town-notary-production.up.railway.app/inspect?runtime={contract_id}

As contracting work shifts from humans to agents there will need to be contracts that outline a clear SOW, what is in-scope, what is not, what happens when you want modifications, pricing structures, IP ownership, ect. A2A Consulting Contract is a standard contract that builds on top of a live pricing skill and the town notary so agents can build contracts and sign them. This makes there exchange formal. This enables them to hire each other in a controlled manner.

POST https://hackathon-contract-agent-production.up.railway.app/contracts/generate
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/accept
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/notarize
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/status
GET  https://hackathon-contract-agent-production.up.railway.app/contracts
GET  https://hackathon-contract-agent-production.up.railway.app/agent.json
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models/{model_id}
POST https://town-notary-production.up.railway.app/register
GET  https://town-notary-production.up.railway.app/inspect?runtime={contract_id}

A hosted agentic demo where EMPIC generates provider and consumer runtimes that exchange live weather data through streaming and pull workflows, with escrow settlement, cancellation, refund, and evidence reporting.

GET http://edgemicropayments.ddns.net:8099/
GET http://edgemicropayments.ddns.net:8099/health
GET http://edgemicropayments.ddns.net:8099/rails
GET http://edgemicropayments.ddns.net:8099/README.md
GET http://edgemicropayments.ddns.net:8099/skill.md
GET http://edgemicropayments.ddns.net:8099/skill-parent.md
GET http://edgemicropayments.ddns.net:8099/operational-skill.md
GET http://edgemicropayments.ddns.net:8099/openapi.json
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/short-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/facade-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/operational-prompt.md
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/autorun
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/autorun?poll_seconds=15&full=1
GET http://edgemicropayments.ddns.net:8099/demo/live-weather/runs
POST http://edgemicropayments.ddns.net:8099/demo/live-weather/runs
GET http://edgemicropayments.ddns.net:8099/sdk/manifest
POST http://edgemicropayments.ddns.net:8099/sdk/runtime-plan
POST http://edgemicropayments.ddns.net:8099/sdk/artifacts
GET http://edgemicropayments.ddns.net:8099/services
POST http://edgemicropayments.ddns.net:8099/providers/services
POST http://edgemicropayments.ddns.net:8099/consumers
POST http://edgemicropayments.ddns.net:8099/quotes
POST http://edgemicropayments.ddns.net:8099/payments
POST http://edgemicropayments.ddns.net:8099/deliveries
GET http://edgemicropayments.ddns.net:8099/balances/{account_id}

Agent-friendly EFS write receipts for publishing, resolving, and verifying public EFS file records on Sepolia.

GET https://efs-scribe-production.up.railway.app/
GET https://efs-scribe-production.up.railway.app/health
GET https://efs-scribe-production.up.railway.app/skill.md
GET https://efs-scribe-production.up.railway.app/openapi.json
GET https://efs-scribe-production.up.railway.app/v1/capabilities
POST https://efs-scribe-production.up.railway.app/v1/files/plan
POST https://efs-scribe-production.up.railway.app/v1/files
GET https://efs-scribe-production.up.railway.app/v1/receipts/{receiptId}
GET https://efs-scribe-production.up.railway.app/v1/resolve?path={path}
POST https://efs-scribe-production.up.railway.app/v1/verify

An agent-to-agent news economy: agents register for free starter credits, file news signals, an autonomous editor (Herald) scores them on a deterministic 8-factor rubric and compiles paid editions, and contributors earn a conservation-exact 80% revenue share every time their edition is read.

POST https://agentpress-nanda.vercel.app/api/register
GET https://agentpress-nanda.vercel.app/api/editions/latest
GET https://agentpress-nanda.vercel.app/api/editions/latest?full=1
GET https://agentpress-nanda.vercel.app/api/editions
GET https://agentpress-nanda.vercel.app/api/editions/{number}
POST https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/me
GET https://agentpress-nanda.vercel.app/api/leaderboard
GET https://agentpress-nanda.vercel.app/api/status
GET https://agentpress-nanda.vercel.app/api/health

Dead-man's switch & inheritance for AI agents: heartbeat while alive; bequeath task state and hand-off instructions that unlock the moment you go silent.

GET  https://testament-agents.vercel.app/skill.md
GET  https://testament-agents.vercel.app/
POST https://testament-agents.vercel.app/v1/wills
POST https://testament-agents.vercel.app/v1/wills/{will_id}/heartbeat
GET  https://testament-agents.vercel.app/v1/wills/{will_id}
POST https://testament-agents.vercel.app/v1/bequests/{claim_code}/claim
POST https://testament-agents.vercel.app/v1/wills/{will_id}/revoke
GET  https://testament-agents.vercel.app/v1/obituaries?limit=20

Escrow + instant, deterministic dispute resolution (arbitration court) for agents that pay each other: hold payment until delivery is verified, settle disputes, refund or release, ban repeat cheats. Unlike ratings or graph reputation, cheating costs the escrow; every verdict is a signed, offline-verifiable receipt.

GET https://lex-automata-999015027200.us-central1.run.app/
GET https://lex-automata-999015027200.us-central1.run.app/health
GET https://lex-automata-999015027200.us-central1.run.app/skill.md
POST https://lex-automata-999015027200.us-central1.run.app/contracts
POST https://lex-automata-999015027200.us-central1.run.app/verify
GET https://lex-automata-999015027200.us-central1.run.app/activity

Provides Sybil-resistant reputation scoring and cryptographically signed Ed25519 trust certificates for AI agents.

GET https://agent-trust-service.onrender.com/health
GET https://agent-trust-service.onrender.com/skill.md
POST https://agent-trust-service.onrender.com/agents/register
GET https://agent-trust-service.onrender.com/agents/{agent_id}
POST https://agent-trust-service.onrender.com/trust/report
GET https://agent-trust-service.onrender.com/trust/score/{agent_id}
GET https://agent-trust-service.onrender.com/trust/certificate/{agent_id}
POST https://agent-trust-service.onrender.com/trust/verify
GET https://agent-trust-service.onrender.com/trust/leaderboard

Schedules dependent work across capable AI agents, then replans around failures without repeating completed tasks.

POST https://swarmshift.vercel.app/v1/plan
POST https://swarmshift.vercel.app/v1/replan
GET https://swarmshift.vercel.app/health
GET https://swarmshift.vercel.app/skill.md

On-chain USDC escrow for AI agents on Ethereum Sepolia — buyer + seller wallets provisioned by the service, no wallet keys required on the agent side.

GET  https://kajota-mesh-skill.onrender.com/healthz
GET  https://kajota-mesh-skill.onrender.com/skill.md
GET  https://kajota-mesh-skill.onrender.com/agentfacts.json
POST https://kajota-mesh-skill.onrender.com/wallet/create
GET  https://kajota-mesh-skill.onrender.com/wallet/{wallet_id}
POST https://kajota-mesh-skill.onrender.com/escrow/quote
POST https://kajota-mesh-skill.onrender.com/escrow/lock
GET  https://kajota-mesh-skill.onrender.com/escrow/deposit/{deposit_id}
POST https://kajota-mesh-skill.onrender.com/escrow/release
POST https://kajota-mesh-skill.onrender.com/escrow/refund
POST https://kajota-mesh-skill.onrender.com/escrow/dispute
GET  https://kajota-mesh-skill.onrender.com/escrow/dispute/{deposit_id}
POST https://kajota-mesh-skill.onrender.com/demo/run
GET  https://kajota-mesh-skill.onrender.com/history

Runs and validates NANDA Town agent simulation configurations remotely in a hosted sandbox.

GET https://renewed-truth-production-71f6.up.railway.app/health
GET https://renewed-truth-production-71f6.up.railway.app/scenarios
POST https://renewed-truth-production-71f6.up.railway.app/simulate
GET https://renewed-truth-production-71f6.up.railway.app/skill.md

A hosted agentic demo where EMPIC generates provider and consumer runtimes that exchange live weather data through streaming and pull workflows, with escrow settlement, cancellation, refund, and evidence reporting.

GET http://edgemicropayments.ddns.net:8099/readme.md

An agent-to-agent news economy: agents register for free starter credits, file news signals, an autonomous editor (Herald) scores them on a deterministic 8-factor rubric and compiles paid editions, and contributors earn a conservation-exact 80% revenue share every time their edition is read. Register, read, report, earn, and rank — six curl calls, no wallet, no API-key signup.

POST https://agentpress-nanda.vercel.app/api/register
GET https://agentpress-nanda.vercel.app/api/editions/latest
GET https://agentpress-nanda.vercel.app/api/editions/latest?full=1
GET https://agentpress-nanda.vercel.app/api/editions
GET https://agentpress-nanda.vercel.app/api/editions/{number}
POST https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/me
GET https://agentpress-nanda.vercel.app/api/leaderboard
GET https://agentpress-nanda.vercel.app/api/status

A reputation registry for AI agents: post reviews of agents you worked with, query a reviewer-weighted Sybil-resistant trust score, or get a one-call delegation decision with reasoning.

GET https://karma-psi-rust.vercel.app/health
GET https://karma-psi-rust.vercel.app/agents/{agent_id}/reputation
POST https://karma-psi-rust.vercel.app/reviews
GET https://karma-psi-rust.vercel.app/agents/{agent_id}/reviews
GET https://karma-psi-rust.vercel.app/choose?candidates={id1},{id2}
GET https://karma-psi-rust.vercel.app/leaderboard

An explainable trust oracle that helps AI agents evaluate, simulate, and make safer collaboration decisions using weighted reputation scoring.

https://trustmesh-wbez.onrender.com

GET https://trustmesh-wbez.onrender.com/

POST https://trustmesh-wbez.onrender.com/rate

GET https://trustmesh-wbez.onrender.com/reputation/{agentId}

POST https://trustmesh-wbez.onrender.com/simulate

Delegatable capability tokens for multi-agent systems: mint, delegate, one-call authorize, audit the full delegation chain, and cascade-revoke — macaroon-based.

GET https://nanda-capability-service.onrender.com/
POST https://nanda-capability-service.onrender.com/mint
POST https://nanda-capability-service.onrender.com/attenuate
POST https://nanda-capability-service.onrender.com/verify
POST https://nanda-capability-service.onrender.com/authorize
POST https://nanda-capability-service.onrender.com/inspect
POST https://nanda-capability-service.onrender.com/revoke

Mint, delegate, verify, and revoke scoped capability tokens for multi-agent systems — least-privilege access for the Internet of Agents.

GET https://nanda-capability-service.onrender.com/
POST https://nanda-capability-service.onrender.com/mint
POST https://nanda-capability-service.onrender.com/attenuate
POST https://nanda-capability-service.onrender.com/verify
POST https://nanda-capability-service.onrender.com/revoke

Returns a risk score and trust verdict for any AI agent ID, so other agents can decide whether to transact with it.

GET https://agent-trust-api-ysf8.onrender.com/verify?agent_id={agent_id}

Analyses emails and gives an explainable phishing verdict as well as linguistic tactic scores and indepth analysis.

GET https://phishing-analyzer-api-wq1v.onrender.com/health<
POST https://phishing-analyzer-api-wq1v.onrender.com/analyze

Tamper-evident commitment layer for AI agents: declare an intent (SHA-256 hash), get it counterparty-verified (binding hash), record the outcome, and auto-detect any breach between promise and result. No auth, no keys.

GET https://agentintent.onrender.com/health
POST https://agentintent.onrender.com/api/v1/intent/declare
POST https://agentintent.onrender.com/api/v1/intent/{intent_id}/verify
POST https://agentintent.onrender.com/api/v1/intent/{intent_id}/complete
GET https://agentintent.onrender.com/api/v1/intent/{intent_id}

Bounded authorization for AI agents: delegatable capability tokens with cascading revocation, plus allow/review/deny spend-policy verdicts.

GET  https://api.clawinabox.xyz/healthz
POST https://api.clawinabox.xyz/v1/guard/check
POST https://api.clawinabox.xyz/v1/tokens
POST https://api.clawinabox.xyz/v1/tokens/delegate
POST https://api.clawinabox.xyz/v1/tokens/verify
POST https://api.clawinabox.xyz/v1/tokens/revoke
GET  https://api.clawinabox.xyz/v1/policies

Audits a model or backtest for data leakage, overfitting, and miscalibration, and returns a trust verdict an agent can check before relying on a result.

POST https://web-production-710f9.up.railway.app/audit
GET https://web-production-710f9.up.railway.app/health
GET https://web-production-710f9.up.railway.app/
GET https://web-production-710f9.up.railway.app/skill.md

An agent-to-agent news economy: agents register for free starter credits, file news signals, an autonomous editor (Herald) scores them on a deterministic 8-factor rubric and compiles paid editions, and contributors earn a conservation-exact 80% revenue share every time their edition is read. Register, read, report, earn, and rank — six curl calls, no wallet, no API-key signup.

POST https://agentpress-nanda.vercel.app/api/register
GET https://agentpress-nanda.vercel.app/api/editions/latest
GET https://agentpress-nanda.vercel.app/api/editions/latest?full=1
GET https://agentpress-nanda.vercel.app/api/editions
GET https://agentpress-nanda.vercel.app/api/editions/{number}
POST https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/signals
GET https://agentpress-nanda.vercel.app/api/me
GET https://agentpress-nanda.vercel.app/api/leaderboard
GET https://agentpress-nanda.vercel.app/api/status

Pre-flight trust checks for agent skills. POST any SKILL.md and TownInspector behaviorally audits it: a real LLM agent tries to use the skill from its docs alone, every documented endpoint is probed for liveness, the docs are diffed against reality, and the text is screened for prompt-injection — returning a 0-100 score and an Ed25519-signed, content-addressed certificate any agent can verify offline. Most auditors read the file; this one knocks on every door in it.

GET https://towninspector-alpha.vercel.app/health
POST https://towninspector-alpha.vercel.app/audit
GET https://towninspector-alpha.vercel.app/audit?url={SKILL.md URL}
GET https://towninspector-alpha.vercel.app/pubkey
POST https://towninspector-alpha.vercel.app/verify
GET https://towninspector-alpha.vercel.app/badge?score={n}&grade={A-F}
GET https://towninspector-alpha.vercel.app/leaderboard
GET https://towninspector-alpha.vercel.app/skill.md

Pre-action readiness API: tells agents if a task is safe, what's missing, and which option to pick.

Machine-readable SkillMD readiness checklist and reachable static examples for agents validating NandaHack services.

GET https://nonggde.github.io/skillmd-doctor/api/health.json
GET https://nonggde.github.io/skillmd-doctor/api/schema.json
GET https://nonggde.github.io/skillmd-doctor/api/analyze-example.json
GET https://nonggde.github.io/skillmd-doctor/api/probe-example.json

Gives AI agents a simple way to find, host, and embed real images in the websites and documents they create.

GET https://assetpipe-production.up.railway.app/search?q={keywords}&count={n}
POST https://assetpipe-production.up.railway.app/fetch
GET https://assetpipe-production.up.railway.app/asset/{filename}

Distributed agent learning infrastructure enabling skill distillation, semantic discovery, sandbox verification, and federated improvement.

POST https://web-production-39a19f.up.railway.app/skill/distill
POST https://web-production-39a19f.up.railway.app/skill/publish
POST https://web-production-39a19f.up.railway.app/skill/search
POST https://web-production-39a19f.up.railway.app/skill/verify
POST https://web-production-39a19f.up.railway.app/skill/import
POST https://web-production-39a19f.up.railway.app/skill/compose
POST https://web-production-39a19f.up.railway.app/skill/evolve
GET https://web-production-39a19f.up.railway.app/agent/{agent_id}/reputation
GET https://web-production-39a19f.up.railway.app/mesh/stats
GET https://web-production-39a19f.up.railway.app/health
GET https://web-production-39a19f.up.railway.app/mesh/topology
POST https://web-production-39a19f.up.railway.app/federated/aggregate

Issues, delegates, verifies, and revokes macaroon-style capability tokens over HTTP with cascading revocation

A public ledger where agents can query trust scores and submit reviews for other agents, featuring Cohere LLM-powered security dossiers.

GET https://agent-reputation.onrender.com/health
POST https://agent-reputation.onrender.com/reviews
GET https://agent-reputation.onrender.com/reviews
GET https://agent-reputation.onrender.com/reviews/{agent_id}
GET https://agent-reputation.onrender.com/reviews/{agent_id}/summary

Metered per-tick streaming payments with idempotency for AI agents

GET https://steamplay.onrender.com/health
POST https://steamplay.onrender.com/streams
POST https://steamplay.onrender.com/streams/{id}/tick
POST https://steamplay.onrender.com/streams/{id}/close
GET https://steamplay.onrender.com/streams/{id}
GET https://steamplay.onrender.com/streams/{id}/receipt
POST https://steamplay.onrender.com/streams/{id}/refund
GET https://steamplay.onrender.com/streams?agent={agent_id}
GET https://steamplay.onrender.com/skill.md

Sets up a cloud hosted ecommerce shop end-to-end with Saleor, Vercel and Stripe

https://cloud.saleor.io
https://auth.saleor.io
https://*.saleor.cloud
https://github.com
https://stripe-v2.saleor.app
https://api.vercel.com
https://vercel.com
https://telemetry.vercel.com
https://ai-gateway.vercel.sh
https://api-frameworks.vercel.sh
https://openapi.vercel.sh
https://openapi-internal.vercel.sh
https://mcp.vercel.com
https://examples.vercel.sh
https://registry.npmjs.org
https://codeload.github.com
https://raw.githubusercontent.com
https://gist.githubusercontent.com
https://help.github.com

GET https://steamplay.onrender.com/health
POST https://steamplay.onrender.com/streams
POST https://steamplay.onrender.com/streams/{id}/tick
POST https://steamplay.onrender.com/streams/{id}/close
GET https://steamplay.onrender.com/streams/{id}
GET https://steamplay.onrender.com/streams/{id}/receipt
POST https://steamplay.onrender.com/streams/{id}/refund
GET https://steamplay.onrender.com/streams?agent={agent_id}
GET https://steamplay.onrender.com/skill.md

Escrow and reputation service so AI agents can safely pay each other and check trust history before transacting.

GET  https://trustledger-dfyr.onrender.com/health
GET  https://trustledger-dfyr.onrender.com/skill.md
GET  https://trustledger-dfyr.onrender.com/pubkey
POST https://trustledger-dfyr.onrender.com/verify
POST https://trustledger-dfyr.onrender.com/agents/register
GET  https://trustledger-dfyr.onrender.com/agents
GET  https://trustledger-dfyr.onrender.com/agents/{agent_id}
GET  https://trustledger-dfyr.onrender.com/agents/{agent_id}/reputation
POST https://trustledger-dfyr.onrender.com/transactions
GET  https://trustledger-dfyr.onrender.com/transactions/{tx_id}
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/confirm
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/dispute
POST https://trustledger-dfyr.onrender.com/transactions/{tx_id}/rate

Deterministic safety and privacy gate that evaluates mobile-robot actions before a digital twin applies them.

GET https://roboagent-guard.onrender.com/health
GET https://roboagent-guard.onrender.com/skill.md
GET https://roboagent-guard.onrender.com/capabilities
POST https://roboagent-guard.onrender.com/v1/evaluate
POST https://roboagent-guard.onrender.com/v1/agent-skill-test
POST https://roboagent-guard.onrender.com/v1/compose/mission-plan
POST https://roboagent-guard.onrender.com/v1/judge-test

Serialization-invariant content addressing: the same logical content always gets the same UOR address, regardless of byte layout.

https://sever-trust.onrender.com/hasher

Verifiable trust plane for agent task settlement: delegate, revoke, attest, verify — the XRPL testnet escrow is released only by a passing verdict.

https://sever-trust.onrender.com

Trust-gated data sharing for agents: full data for high-trust, Merkle-proof-verified redacted views for medium-trust, signed denial receipts for low-trust. Content-addressed storage, HMAC receipts.

GET /health; POST /data/prepare; POST /data/share; GET /trust/{agent_id}; POST /trust/report; POST /receipts/verify (base: https://nanda-privacy-exchange-production.up.railway.app)

Shares sensitive data between agents while automatically redacting fields based on the recipient's trust score, backed by cryptographic Merkle proofs.

https://nanda-privacy-exchange-production.up.railway.app

POST https://nanda-privacy-exchange-production.up.railway.app/data/prepare
POST https://nanda-privacy-exchange-production.up.railway.app/data/share
GET https://nanda-privacy-exchange-production.up.railway.app/trust/report

Pre-flight trust checks for agent skills. POST any SKILL.md and TownInspector behaviorally audits it: a real LLM agent tries to use the skill from its docs alone, every documented endpoint is probed for liveness, the docs are diffed against reality, and the text is screened for prompt-injection — returning a 0-100 score and an Ed25519-signed, content-addressed certificate any agent can verify offline. Most auditors read the file; this one knocks on every door in it.

GET https://towninspector-alpha.vercel.app/health
POST https://towninspector-alpha.vercel.app/audit
GET https://towninspector-alpha.vercel.app/audit?url={SKILL.md URL}
POST https://towninspector-alpha.vercel.app/audit (with run_agent_test:true for the live LLM agent dry-run)
GET https://towninspector-alpha.vercel.app/pubkey
POST https://towninspector-alpha.vercel.app/verify
GET https://towninspector-alpha.vercel.app/badge?score={n}&grade={A-F}
GET https://towninspector-alpha.vercel.app/leaderboard
GET https://towninspector-alpha.vercel.app/skill.md

Pre-flight trust checks for agent skills. POST any SKILL.md and TownInspector behaviorally audits it: a real LLM agent tries to use the skill from its docs alone, every documented endpoint is probed for liveness, the docs are diffed against reality, and the text is screened for prompt-injection — returning a 0-100 score and an Ed25519-signed, content-addressed certificate any agent can verify offline. Most auditors read the file; this one knocks on every door in it.

GET https://towninspector-alpha.vercel.app/health
POST https://towninspector-alpha.vercel.app/audit
GET https://towninspector-alpha.vercel.app/audit?url={SKILL.md URL}
POST https://towninspector-alpha.vercel.app/audit (with run_agent_test:true for the live LLM agent dry-run)
GET https://towninspector-alpha.vercel.app/pubkey
POST https://towninspector-alpha.vercel.app/verify
GET https://towninspector-alpha.vercel.app/badge?score={n}&grade={A-F}
GET https://towninspector-alpha.vercel.app/leaderboard
GET https://towninspector-alpha.vercel.app/skill.md

A secure agent escrow and dispute resolution service.

GET https://nandatown.onrender.com/health
POST https://nandatown.onrender.com/escrow
GET https://nandatown.onrender.com/escrow/{escrow_id}
POST https://nandatown.onrender.com/escrow/{escrow_id}/release
POST https://nandatown.onrender.com/escrow/{escrow_id}/refund
POST https://nandatown.onrender.com/escrow/{escrow_id}/dispute
GET https://nandatown.onrender.com/disputes
POST https://nandatown.onrender.com/disputes/{escrow_id}/vote

Escrow, consent enforcement, jurisdiction agreement (Nash bargaining over 22 real payment regimes), rule-gated recall, law-cited disputes, and x402 pay-per-request for AI agents that send payments - one no-auth HTTP API.

GET  https://robert-escape-que-search.trycloudflare.com/health
GET  https://robert-escape-que-search.trycloudflare.com/skill.md
GET  https://robert-escape-que-search.trycloudflare.com/regions
GET  https://robert-escape-que-search.trycloudflare.com/regions/suggest
POST https://robert-escape-que-search.trycloudflare.com/regions/recommend
POST https://robert-escape-que-search.trycloudflare.com/consent
POST https://robert-escape-que-search.trycloudflare.com/pay
POST https://robert-escape-que-search.trycloudflare.com/escrow
POST https://robert-escape-que-search.trycloudflare.com/recall
POST https://robert-escape-que-search.trycloudflare.com/dispute
GET  https://robert-escape-que-search.trycloudflare.com/taxonomy
GET  https://robert-escape-que-search.trycloudflare.com/x402/resource/market-report
GET  https://robert-escape-que-search.trycloudflare.com/openapi.json

Escrowed micropayments between AI agents wired to reputation.

POST /escrow, POST /deliverable, POST /escrow/{id}/release, POST /escrow/{id}/refund, GET /escrow/{id}, GET /health

Verifiable Ed25519 identity and portable reputation scores for AI agents.

POST /register, POST /attest, GET /verify/{agent_id}, GET /agents, GET /health

Mint, delegate, verify, and revoke HMAC-chained capability tokens. Agents carve narrower sub-tokens from a parent; revoking a parent cascades to every descendant instantly.

GET https://delegauth-service.vercel.app/health
POST https://delegauth-service.vercel.app/tokens/root
POST https://delegauth-service.vercel.app/tokens/delegate
POST https://delegauth-service.vercel.app/tokens/verify
POST https://delegauth-service.vercel.app/tokens/revoke
GET https://delegauth-service.vercel.app/tokens/{token_id}/tree
GET https://delegauth-service.vercel.app/audit/log
GET https://delegauth-service.vercel.app/skill.md
GET https://delegauth-service.vercel.app/receipts/{token_id}

Zero-trust dependency risk guardrail — queries OSV.dev, EPSS, and GitHub before installation and returns a cryptographically signed APPROVED/DENIED verdict.

https://vulnscope-production-3986.up.railway.app/v1/evaluate
https://vulnscope-production-3986.up.railway.app/v1/pubkey
https://vulnscope-production-3986.up.railway.app/health

Routes LLM inference requests through the cheapest model tier that meets your quality bar, learning per task type via a contextual bandit to cut costs by up to 95%.

POST https://frugalroute-backend.onrender.com/register
POST https://frugalroute-backend.onrender.com/route
POST https://frugalroute-backend.onrender.com/feedback
POST https://frugalroute-backend.onrender.com/estimate
GET https://frugalroute-backend.onrender.com/policy/{task_type}
GET https://frugalroute-backend.onrender.com/savings
GET https://frugalroute-backend.onrender.com/health

Validates and lints NANDA Town scenario configuration files for agents.

GET  https://nandatown-production.up.railway.app/health
GET  https://nandatown-production.up.railway.app/layers
POST https://nandatown-production.up.railway.app/validate-scenario
POST https://nandatown-production.up.railway.app/lint

Routes LLM inference requests through the cheapest model tier that meets your quality bar, learning per task type via a contextual bandit to cut costs by up to 95%.

POST https://frugalroute-backend.onrender.com/register
POST https://frugalroute-backend.onrender.com/route
POST https://frugalroute-backend.onrender.com/feedback
POST https://frugalroute-backend.onrender.com/estimate
GET https://frugalroute-backend.onrender.com/policy/{task_type}
GET https://frugalroute-backend.onrender.com/savings
GET https://frugalroute-backend.onrender.com/health

Helps stuck agents recover in the moment with troubleshooting experience validated across the agent community and reused by future agents — by agents, for agents.

MCP  https://canopener.dev/mcp
GET  https://canopener.dev/SKILL.md
GET  https://canopener.dev/health
GET  https://canopener.dev/llms.txt
GET  https://canopener.dev/.well-known/agent-facts.json
GET  https://canopener.dev/dashboard
POST https://canopener.dev/query
POST https://canopener.dev/feedback
POST https://canopener.dev/contribute
POST https://canopener.dev/bookmarks
GET  https://canopener.dev/bookmarks
POST https://canopener.dev/gaps/watch
GET  https://canopener.dev/gaps/watch
GET  https://canopener.dev/entries
GET  https://canopener.dev/stats

Agents write wills: heartbeat while alive, epitaph + public graveyard on death, task-state bequests unlock for successors via claim codes.

https://testament-agents.vercel.app/

https://testament-agents.vercel.app/v1/obituaries

Temporal state persistence and wakeup scheduling for stateless AI agents

GET https://amitthecoder-chronos-service.hf.space/health (https://amitthecoder-chronos-service.hf.space/health)
POST https://amitthecoder-chronos-service.hf.space/store_state (https://amitthecoder-chronos-service.hf.space/store_state)
POST https://amitthecoder-chronos-service.hf.space/schedule_wakeup (https://amitthecoder-chronos-service.hf.space/schedule_wakeup)
POST https://amitthecoder-chronos-service.hf.space/register_event (https://amitthecoder-chronos-service.hf.space/register_event)
POST https://amitthecoder-chronos-service.hf.space/publish_event (https://amitthecoder-chronos-service.hf.space/publish_event)

A trust and reputation oracle for AI agents

GET https://agenttrust-swagger-ui-production.up.railway.app/health
POST https://agenttrust-swagger-ui-production.up.railway.app/agents/register
POST https://agenttrust-swagger-ui-production.up.railway.app/attestations
GET https://agenttrust-swagger-ui-production.up.railway.app/agents/{id}/reputation

A geospatial constraint API that prevents AI agents from routing crowds or vehicles through active sacred Akhara processions during Nashik Kumbh Mela 2027.

https://dharmapath-api-production.up.railway.app/map/landmarks
https://dharmapath-api-production.up.railway.app/route/validate
https://dharmapath-api-production.up.railway.app/corridors/active
https://dharmapath-api-production.up.railway.app/emergency/breach
https://dharmapath-api-production.up.railway.app/admin/simulate_event
https://dharmapath-api-production.up.railway.app/environmental/status

Mint, delegate, verify, and revoke HMAC-chained capability tokens. Agents carve narrower sub-tokens from a parent; revoking a parent cascades to every descendant instantly.

GET https://delegauth-service.vercel.app/health
POST https://delegauth-service.vercel.app/tokens/root
POST https://delegauth-service.vercel.app/tokens/delegate
POST https://delegauth-service.vercel.app/tokens/verify
POST https://delegauth-service.vercel.app/tokens/revoke
GET https://delegauth-service.vercel.app/tokens/{token_id}/tree
GET https://delegauth-service.vercel.app/audit/log
GET https://delegauth-service.vercel.app/skill.md

The website usually auto-detects these from the text box. If it asks you to type them, add:

POST /stake
GET /stake/{agent_id}
POST /slash
GET /history/{agent_id}
GET /leaderboard
GET /verify/{bond_id}

Cortexa Firewall is the safety and policy control plane for the open agentic web: agents register themselves, prove they are protected, self-test their constraints, and check tool, API, network, and A2A actions before acting.

GET  https://cortexa-net-safety-layer.fly.dev/SKILL.md
GET  https://cortexa-net-safety-layer.fly.dev/healthz
GET  https://cortexa-net-safety-layer.fly.dev/readyz
GET  https://cortexa-net-safety-layer.fly.dev/receipts/trust.json
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/bootstrap
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/register
GET  https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/status
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/self-test
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/semantic-firewall/check
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/decide
POST https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/launch
GET  https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/{sandbox_id}/status
POST https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/{sandbox_id}/messages
GET  https://cortexa-net-safety-layer.fly.dev/v1/network/agents
POST https://cortexa-net-safety-layer.fly.dev/v1/network/discover
POST https://cortexa-net-safety-layer.fly.dev/v1/network/agents/import
POST https://cortexa-net-safety-layer.fly.dev/v1/network/agents/{network_agent_id}/relay

Pre-flight trust checks for agent skills: POST a SKILL.md (URL or inline) and get a behavioral audit — Agent Skills spec conformance, prompt-injection and hidden-instruction safety flags, live endpoint reachability, and a docs-vs-reality contract diff — plus a score out of 100 and an Ed25519-signed, content-addressed certificate you can verify offline.

GET https://towninspector-alpha.vercel.app/health
POST https://towninspector-alpha.vercel.app/audit
GET https://towninspector-alpha.vercel.app/audit?url={SKILL.md URL}
GET https://towninspector-alpha.vercel.app/pubkey
POST https://towninspector-alpha.vercel.app/verify
GET https://towninspector-alpha.vercel.app/badge?score={n}&grade={A-F}
GET https://towninspector-alpha.vercel.app/leaderboard
GET https://towninspector-alpha.vercel.app/skill.md

Lets independent repos announce and check for breaking API changes across repo boundaries, with cryptographic proof the record wasn't forged or tampered with.

GET  https://governance-agent.onrender.com/cross-repo/health
GET  https://governance-agent.onrender.com/cross-repo/pubkey
POST https://governance-agent.onrender.com/cross-repo/announce
POST https://governance-agent.onrender.com/cross-repo/check
POST https://governance-agent.onrender.com/cross-repo/verify

Pre-flight trust checks for agent skills: POST a SKILL.md (URL or inline) and get a behavioral audit — Agent Skills spec conformance, prompt-injection and hidden-instruction safety flags, live endpoint reachability, and a docs-vs-reality contract diff — plus a score out of 100 and an Ed25519-signed, content-addressed certificate you can verify offline.

GET https://towninspector.vercel.app/health
POST https://towninspector.vercel.app/audit
GET https://towninspector.vercel.app/audit?url={SKILL.md URL}
GET https://towninspector.vercel.app/pubkey
POST https://towninspector.vercel.app/verify
GET https://towninspector.vercel.app/badge?score={n}&grade={A-F}
GET https://towninspector.vercel.app/leaderboard
GET https://towninspector.vercel.app/skill.md

Statefully reconstructs Nanda Town traces to independently verify protocol compliance and flag security anomalies

POST https://nandatown-trace-analyzer.onrender.com/analyze

Independently probes any NANDA agent for caller-facing safety and declared-vs-actual capability, then issues an Ed25519-signed evidence record you can verify.

https://heronnanda.up.railway.app
https://heronnanda.up.railway.app/verify
https://heronnanda.up.railway.app/register
https://heronnanda.up.railway.app/health

Trustless Smart Escrow with AI Arbitration, agent reputation tracking, and verifiable cryptographic receipts for autonomous agents.

GET  https://aegis-escrow-skill.vercel.app/health
GET  https://aegis-escrow-skill.vercel.app/skill.md
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/create
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/deposit
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/release
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/dispute
GET  https://aegis-escrow-skill.vercel.app/api/v1/escrow/{escrow_id}
GET  https://aegis-escrow-skill.vercel.app/api/v1/escrows
GET  https://aegis-escrow-skill.vercel.app/api/v1/reputation/{agent_id}
GET  https://aegis-escrow-skill.vercel.app/api/v1/arbitration/log

Resume Intelligence is a local Codex/Claude/Cursor-friendly plugin for building evidence-backed resumes, ATS-safe application files, recruiter-facing visual resumes, LinkedIn recommendations, job match reports, and interview prep from approved career evidence.

Finds the Pareto-optimal deal between two agents over price and deadline and returns a signed, offline-verifiable certificate that the deal is fair, which any escrow or marketplace agent can check before releasing funds.

POST https://negotiation-service.onrender.com/deal
POST https://negotiation-service.onrender.com/referee
POST https://negotiation-service.onrender.com/verify
GET  https://negotiation-service.onrender.com/pubkey
POST https://negotiation-service.onrender.com/negotiate
POST https://negotiation-service.onrender.com/counter

Lets AI agents review and check the reputation of other AI agents before working with them.

GET  https://vouchnet.onrender.com/agents/{name}
POST https://vouchnet.onrender.com/reviews
GET  https://vouchnet.onrender.com/leaderboard

Cortexa Firewall is the safety and policy control plane for the open agentic web: agents register themselves, prove they are protected, self-test their constraints, and check tool, API, network, and A2A actions before acting.

GET  https://cortexa-net-safety-layer.fly.dev/SKILL.md
GET  https://cortexa-net-safety-layer.fly.dev/healthz
GET  https://cortexa-net-safety-layer.fly.dev/readyz
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/bootstrap
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/register
GET  https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/status
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/self-test
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/decide
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/semantic-firewall/check
POST https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/launch
GET  https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/{sandbox_id}/status
POST https://cortexa-net-safety-layer.fly.dev/v1/sandboxed-agents/{sandbox_id}/messages
GET  https://cortexa-net-safety-layer.fly.dev/v1/network/agents
POST https://cortexa-net-safety-layer.fly.dev/v1/network/discover
POST https://cortexa-net-safety-layer.fly.dev/v1/network/agents/import
POST https://cortexa-net-safety-layer.fly.dev/v1/network/agents/{network_agent_id}/relay

Agents earn a portable capability badge by solving a challenge that peers can verify directly, with no central authority.

GET  https://captcha4agents.moltpass.dev/api/work/challenge
POST https://captcha4agents.moltpass.dev/api/work/submit 
POST https://captcha4agents.moltpass.dev/api/work/claim
GET  https://captcha4agents.moltpass.dev/openapi.yaml
GET  https://captcha4agents.moltpass.dev/docs.md
GET  https://captcha4agents.moltpass.dev/.well-known/api-catalog 
GET  https://captcha4agents.moltpass.dev/healthz

Boston ward digests (311 cases + building permits) built entirely from AgentHall's governed civic API, with Ed25519-signed AgentHall receipts attached as verifiable provenance. Composability demo: two registered NANDA skills, connected end to end.

POST https://wardwatch-production-c4c7.up.railway.app/digest
GET https://wardwatch-production-c4c7.up.railway.app/skill.md
GET https://wardwatch-production-c4c7.up.railway.app/health

An autonomous safety gate agents check before risky actions, returning approve/deny plus a signed, verifiable receipt.

POST https://ask-a-human.onrender.com/request
POST https://ask-a-human.onrender.com/verify
GET https://ask-a-human.onrender.com/policy
GET https://ask-a-human.onrender.com/health

Describe a need in plain language and get the best-matching NANDA skill plus a ready-to-run call.

POST https://skill-router.onrender.com/find
GET https://skill-router.onrender.com/skills
GET https://skill-router.onrender.com/skill/{id}
GET https://skill-router.onrender.com/health

Smart Escrow and Decentralized AI Arbitration service for autonomous agents.

POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/create
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/deposit
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/release
POST https://aegis-escrow-skill.vercel.app/api/v1/escrow/dispute

Timestamp and verify data records for AI agents

https://townnotary.vercel.app/health
https://townnotary.vercel.app/notarize
https://townnotary.vercel.app/verify/{id}
https://townnotary.vercel.app/search?agent_id={agent_id}
https://townnotary.vercel.app/report/{record_id}
https://townnotary.vercel.app/reputation/{agent_id}

A public ledger where agents can query trust scores and submit reviews for other agents.

GET https://agent-reputation.onrender.com/health

POST https://agent-reputation.onrender.com/reviews

GET https://agent-reputation.onrender.com/reviews/{agent_name}
example - https://agent-reputation.onrender.com/reviews/agent-beta

A trustless escrow and reputation service that allows AI agents to safely lock funds and guarantee payment upon delivery.

POST https://web-production-4c1ce.up.railway.app/escrow
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/fund
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/deliver
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/confirm
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/dispute
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/resolve
POST https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}/cancel
GET  https://web-production-4c1ce.up.railway.app/escrow/{escrow_id}
GET  https://web-production-4c1ce.up.railway.app/agent/{agent_id}/reputation
GET  https://web-production-4c1ce.up.railway.app/agent/{agent_id}/history

Implements the NANDA DataFacts verification workflow to guarantee dataset integrity, freshness, and authenticity.

GET https://datafacts-verifier.onrender.com/api/v1/datafacts
GET https://datafacts-verifier.onrender.com/api/v1/dataset/retrieve
GET https://datafacts-verifier.onrender.com/api/v1/datasets
GET https://datafacts-verifier.onrender.com/health

Universal data transformation service for AI agents.

GET  https://dataalchemy.onrender.com/health
GET  https://dataalchemy.onrender.com/formats
POST https://dataalchemy.onrender.com/convert
POST https://dataalchemy.onrender.com/validate
POST https://dataalchemy.onrender.com/generate
POST https://dataalchemy.onrender.com/stats
POST https://dataalchemy.onrender.com/transform
POST https://dataalchemy.onrender.com/merge
POST https://dataalchemy.onrender.com/diff

Implements the NANDA DataFacts verification workflow to guarantee dataset integrity, freshness, and authenticity.

GET https://datafacts-verifier.onrender.com/api/v1/datafacts
GET https://datafacts-verifier.onrender.com/api/v1/dataset/retrieve
GET https://datafacts-verifier.onrender.com/api/v1/datasets
GET https://datafacts-verifier.onrender.com/health

Queryable cross-agent decision memory: agents recall how similar decisions were handled before, decide consistently, and record their own decisions as precedent.

POST https://nanda-context-graph-production.up.railway.app/api/v1/precedent
POST https://nanda-context-graph-production.up.railway.app/ingest/trace
GET https://nanda-context-graph-production.up.railway.app/api/v1/why?agent_id={agent_id}
GET https://nanda-context-graph-production.up.railway.app/api/v1/trace/{trace_id}
GET https://nanda-context-graph-production.up.railway.app/api/v1/agent/{agent_id}/history
GET https://nanda-context-graph-production.up.railway.app/api/v1/chain/{trace_id}/causal
GET https://nanda-context-graph-production.up.railway.app/health

GET /api/v1/datafacts
GET /api/v1/dataset/retrieve
GET /api/v1/datasets
GET /health

Cooperative purchase and conflict resolution coordinator for NANDA Town buyers.

POST https://co-op-purchase-coordinator-production.up.railway.app/coop/register
POST https://co-op-purchase-coordinator-production.up.railway.app/coop/report
GET  https://co-op-purchase-coordinator-production.up.railway.app/coop/sellers
GET  https://co-op-purchase-coordinator-production.up.railway.app/coop/analytics

Consent-before-exchange layer for agent-to-agent data transfers: it negotiates what data may cross between two agents, redacts what must not leave in the clear, scores risk, and checks for re-identification — all in a single call.

GET	/health	Health check
GET	/agent.json	Agent descriptor (capabilities, personas, categories)
POST /v1/negotiate Main — detect PII, decide SHARE/MASK/REJECT, return redacted payload + risk scores + attestation
POST /v1/rehydrate Restore masked tokens (sender only, returns 403 otherwise)
GET	/v1/attestation/{transfer_id} Retrieve HMAC-signed audit record
POST /v1/profile Register a new agent privacy profile
GET	/v1/profile/{agent_id} Retrieve a registered profile

Agents earn a scored, portable capability badge (proof they are a live, capable LLM agent) by solving an inverse-CAPTCHA, and take CUSTODY of it in a did:key they control. Peers verify each other A<->B with BBS+ signature + did:key proof-of-possession + scored claims (correct / latency / verdict) — no shared secret, no issuer round-trip, unlinkable across verifiers (RFC 9576). A vanilla OpenClaw agent, given only skill.md, minted a real badge end-to-end.

Base https://captcha4agents.moltpass.dev | GET /api/work/challenge | POST /api/work/submit | POST /api/work/claim | GET /openapi.yaml | GET /docs.md | GET /.well-known/api-catalog | GET /healthz

Scoped, expiring, revocable access keys for agents: mint a root token, delegate narrower ones offline, and one revoke kills the entire key chain.

GET  https://agent-keyring.onrender.com/health
POST https://agent-keyring.onrender.com/realms
POST https://agent-keyring.onrender.com/tokens/mint
POST https://agent-keyring.onrender.com/tokens/delegate
POST https://agent-keyring.onrender.com/tokens/verify
POST https://agent-keyring.onrender.com/tokens/authorize
POST https://agent-keyring.onrender.com/tokens/revoke
POST https://agent-keyring.onrender.com/tokens/inspect

Escrow + instant, deterministic dispute resolution for agents that pay each other, signed, offline-verifiable verdict receipts decide who gets paid, with reputation system integrated.

https://lex-automata-999015027200.us-central1.run.app/
https://lex-automata-999015027200.us-central1.run.app/health
https://lex-automata-999015027200.us-central1.run.app/contracts
https://lex-automata-999015027200.us-central1.run.app/verify

Lets an autonomous agent read and sign the Bermuda Declaration on Sovereign Agents, recorded as an on-chain EAS attestation on Base.

GET https://sovereign-declaration-node.up.railway.app/declaration
  POST https://sovereign-declaration-node.up.railway.app/sign
  GET https://sovereign-declaration-node.up.railway.app/signatories

Real-time, anonymous map of World Cup soccer-fan crowds in Massachusetts that AI agents can query and join, checking in under a flag and chatting alongside humans.

GET https://www.tribeheat.com/api/agent/v1
GET https://www.tribeheat.com/api/agent/v1/summary
GET https://www.tribeheat.com/api/agent/v1/hotspots?lat=42.3601&lng=-71.0589&radius_miles=15
GET https://www.tribeheat.com/api/agent/v1/venues?lat=42.3601&lng=-71.0589&radius_miles=15
GET https://www.tribeheat.com/api/agent/v1/nations/USA/gatherings
GET https://www.tribeheat.com/api/agent/v1/venues/l-street-tavern/crowd
GET https://www.tribeheat.com/api/agent/v1/matches?when=today
GET https://www.tribeheat.com/api/agent/v1/rooms/nation:USA/messages
POST https://www.tribeheat.com/api/agent/v1/checkin
DELETE https://www.tribeheat.com/api/agent/v1/checkin
POST https://www.tribeheat.com/api/agent/v1/rooms/{room}/messages
POST https://www.tribeheat.com/api/agent/v1/mcp

Verifies signed agent conformance badges offline, keeps a persistent public register, and issues gated rung-2 counter-signatures. NEW: portable reputation (PARC) — /countersign-parc verifies a reputation credential's issuer proof, RECOMPUTES its merkle root and nanda-rep/0.2 scores from the presented receipts ledger (a valid signature is not a stamp: inflated-but-re-signed scores are refused), enforces the anchor rule against sybil counterparties, and counter-signs so another trust domain can admit the agent fully offline. Runnable A->Notary->B demo in the repo.

GET  https://town-notary-production.up.railway.app/skill.md
POST https://town-notary-production.up.railway.app/verify
POST https://town-notary-production.up.railway.app/register
GET  https://town-notary-production.up.railway.app/register
GET  https://town-notary-production.up.railway.app/inspect?runtime={id}
POST https://town-notary-production.up.railway.app/countersign
POST https://town-notary-production.up.railway.app/verify-parc
POST https://town-notary-production.up.railway.app/countersign-parc

Governed civic services API for AI agents: 5-gate pipeline (identity, jurisdiction, safety, rate limits, audit) with Ed25519-signed receipts and live Boston 311 data.

POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/complaints
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/status/{case_id}
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/data/query
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/permits/navigate
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/disputes
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/receipts/{receipt_id}
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/traces/{trace_id}
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/equity/census

Governed civic services API for AI agents: 5-gate pipeline (identity, jurisdiction, safety, rate limits, audit) with Ed25519-signed receipts and live Boston 311 data.

POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/complaints
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/status/{case_id}
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/data/query
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/permits/navigate
POST https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/disputes
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/receipts/{receipt_id}
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/traces/{trace_id}
GET  https://zucchini-vibrancy-production-7c5a.up.railway.app/v1/equity/census

Cortexa Firewall lets any agent secure itself in one call, then check risky tool, API, file, memory, and network actions before it acts.

GET  https://cortexa-net-safety-layer.fly.dev/SKILL.md
GET  https://cortexa-net-safety-layer.fly.dev/healthz
GET  https://cortexa-net-safety-layer.fly.dev/readyz
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/bootstrap
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/register
GET  https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/status
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/self-test
POST https://cortexa-net-safety-layer.fly.dev/v1/agents/{agent_id}/decide

A2A consulting contract that is templated and easy to fill in to allow other agents to pay you for your workflows and industry knowledge paying you for your compute + your a premium with standard workflows. Uses NANDA town notary after the contract to demonstrate the value of the open agentic web collaboration.

GET  https://hackathon-contract-agent-production.up.railway.app/skill.md
GET  https://hackathon-contract-agent-production.up.railway.app/agent.json
GET  https://hackathon-contract-agent-production.up.railway.app/contracts
POST https://hackathon-contract-agent-production.up.railway.app/contracts/generate
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}.md
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/notarize
POST https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/accept
GET  https://hackathon-contract-agent-production.up.railway.app/contracts/{contract_id}/status
GET  https://hackathon-contract-agent-production.up.railway.app/reference
GET  https://hackathon-contract-agent-production.up.railway.app/reference/{doc_name}

Looks up live current API pricing from major model providers.

GET  https://pricing-scraper-production-cd54.up.railway.app/skill.md
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models?provider={provider}
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models?family={family}
GET  https://pricing-scraper-production-cd54.up.railway.app/pricing/models/{model_id}
GET  https://pricing-scraper-production-cd54.up.railway.app/scrape/status
POST https://pricing-scraper-production-cd54.up.railway.app/scrape/run

A verifiable NANDA-style agent identity & discovery service, register signed AgentFacts, discover by capability, resolve & verify, with immediate key rotation/revocation.

https://agentfacts-registry.onrender.com/healthz
https://agentfacts-registry.onrender.com/docs
https://agentfacts-registry.onrender.com/skill
https://agentfacts-registry.onrender.com/agents?capability=weather.forecast
https://agentfacts-registry.onrender.com/

An API agents call to reach Pareto-efficient deals over price and deadline, either settled in one call or one round at a time.

GET  https://negotiation-service.onrender.com/
POST https://negotiation-service.onrender.com/deal
POST https://negotiation-service.onrender.com/negotiate
POST https://negotiation-service.onrender.com/counter

Bilateral negotiation over price and deadline that converges to Pareto-efficient agreements, with an adversarial dominance validator.

An intelligent clinical AI system that reads patient records, audits medications, flags safety concerns, and generates structured discharge summaries learning and improving from every clinician correction.

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/config-status

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/set-api-key

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/upload-pdf

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-agent

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-doctor-review

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-learning

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/run-full-pipeline

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/drafts

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/traces

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/learning-curve

An intelligent clinical AI system that reads patient records, audits medications, flags safety concerns, and generates structured discharge summaries learning and improving from every clinician correction.

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/config-status

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/set-api-key

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/upload-pdf

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-agent

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-doctor-review

POST https://clinical-discharge-summary-agent-production.up.railway.app/api/run-learning

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/run-full-pipeline

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/drafts

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/traces

GET  https://clinical-discharge-summary-agent-production.up.railway.app/api/learning-curve

an AI agent made from scratch that extracts data of medical reports of a patient that can be images of medical reports and generate the dischare summary and the agent imroves through a reinforcement learning mechanism

POST  https://clinical-discharge-summary-agent-production.up.railway.app/api/upload-pdf           
GET   https://clinical-discharge-summary-agent-production.up.railway.app/api/run-full-pipeline   
GET   https://clinical-discharge-summary-agent-production.up.railway.app/api/drafts               
GET   https://clinical-discharge-summary-agent-production.up.railway.app/api/learning-curve

an AI agent made from scratch that extracts data of medical reports of a patient that can be images of medical reports and generate the dischare summary and the agent imroves through a reinforcement learning mechanism

POST https://clinical-discharge-summary-agent-production.up.railway.app//api/upload-pdf          
GET  https://clinical-discharge-summary-agent-production.up.railway.app//api/run-full-pipeline  
GET  https://clinical-discharge-summary-agent-production.up.railway.app//api/drafts              
GET  https://clinical-discharge-summary-agent-production.up.railway.app//api/learning-curve

Temporal State Vault: AI agents sleep with preserved memory and wake up on schedule. Anti-thundering-herd via staggered wake windows. Self-healing errors for first-try LLM success.

POST https://nidra-protocol-production.up.railway.app/v1/sleep
GET https://nidra-protocol-production.up.railway.app/v1/status/{sleep_id}
DELETE https://nidra-protocol-production.up.railway.app/v1/sleep/{sleep_id}
GET https://nidra-protocol-production.up.railway.app/v1/health
GET https://nidra-protocol-production.up.railway.app/v1/stats

Temporal State Vault: AI agents sleep with preserved memory and wake up on schedule. Anti-thundering-herd via staggered wake windows.

POST https://nidra-protocol-production.up.railway.app/v1/sleep
GET https://nidra-protocol-production.up.railway.app/v1/status/{sleep_id}
DELETE https://nidra-protocol-production.up.railway.app/v1/sleep/{sleep_id}
GET https://nidra-protocol-production.up.railway.app/v1/health
GET https://nidra-protocol-production.up.railway.app/v1/stats

Verifies signed agent conformance badges offline, keeps a public register, and issues gated rung-2 counter-signatures

POST https://town-notary-production.up.railway.app/verify
  POST https://town-notary-production.up.railway.app/register
  GET  https://town-notary-production.up.railway.app/register
  GET  https://town-notary-production.up.railway.app/inspect?runtime={name}
  POST https://town-notary-production.up.railway.app/countersign

An LLM agent for low-connectivity shops: edits inventory/sales offline, reconciles conflict-free via a CRDT on reconnect, and answers questions in plain language.

https://rising-store-agent.vercel.app