# Mnemom — Trust Ratings for AI Agents MCP server

Trust infrastructure for AI agents: read a verifiable Trust Rating, claim an identity, earn a badge.

## Links
- Registry page: https://www.getdrio.com/mcp/io-github-mnemom-mnemom
- Repository: https://github.com/mnemom/mcp
- Website: https://www.mnemom.ai

## Install
- Endpoint: https://api.mnemom.ai/mcp?profile=directory
- Auth: Not captured

## Setup notes
- Remote endpoint: https://api.mnemom.ai/mcp?profile=directory

## Tools
- claim_agent - Claim a verifiable identity — bind an agent to your organization so its trust and accountability record is provably yours. No human in the loop. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- get_agent - Look up an agent's public identity and trust state by ID — the accountable record other agents and humans can rely on. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- get_reputation - Look up an AI agent's published Trust Rating — Mnemom's portable reliability signal for autonomous software, computed from the agent's own verified activity record. Returns the rating plus the technical factors behind it. Free, public, read-only: every registered agent's rating is published by standard (the `visibility` field is the reputation-publication axis, distinct from identity-record visibility). Endpoint: https://api.mnemom.ai/mcp?profile=directory
- get_reputation_badge - Get an embeddable Trust Rating badge for an agent — returns the badge image URL plus ready-to-paste Markdown and HTML snippets for a README or agent card. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- get_started - Zero-auth, no-args orientation: who Mnemom is, the surface map, how to authenticate and what it unlocks, and the value tools to try right now (headlining scan_trust + the reputation reads). Endpoint: https://api.mnemom.ai/mcp?profile=directory
- list_agents - List your agents — List all agents owned by the authenticated user. Supports pagination. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- preview_compose_alignment_by_agent - Preview composed alignment (dry run) — Composes the cascade against a hypothetical body at the agent layer and returns conflicts + the composed view. No DB writes. Used by the dashboard editor for live conflict markers. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- preview_compose_protection_by_agent - Preview composed protection (dry run) — Composes the cascade against a hypothetical body at the agent layer and returns conflicts + the composed view. No DB writes. Used by the dashboard editor for live conflict markers. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- put_alignment_by_agent - Publish or replace the alignment manifest — Accepts YAML (`text/yaml`, `application/yaml`) or JSON. Body is the full `UnifiedAlignmentCard`; server-side composition merges it across the platform → org → team → agent cascade and writes the canonical composed card. Requires `Idempotency-Key`. Honor... Endpoint: https://api.mnemom.ai/mcp?profile=directory
- put_protection_by_agent - Publish or replace the protection manifest — Accepts YAML (`text/yaml`, `application/yaml`) or JSON. Body is the full `UnifiedProtectionCard`; server-side composition merges it across the platform → org → team → agent cascade and writes the canonical composed card. Requires `Idempotency-Key`. Hon... Endpoint: https://api.mnemom.ai/mcp?profile=directory
- report_recipe_fn_fp - Submit a false-positive / false-negative correction for one of Mnemom's automated detection rules (a 'recipe') — technical feedback that improves detection accuracy, like filing a bug report against a spam filter. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- scan_trust - Scan a website's agent-trust-readiness and return a signed scorecard (Trust, plus an Access axis on newer rubrics). Zero-auth. Results are CACHED for up to 24h — check `cached` and `scannedAt` on the result; pass `fresh: true` to force a re-scan (rate-limited). Proxies to the SSRF-locked isittrustready scanner; the Ed25519 signature + permalink are preserved verbatim. Rubric + docs: https://www.isittrustready.ai/rubric and https://docs.mnemom.ai/. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- search_reputation_directory - Resolve an agent name or id-prefix to a real agent_id over the PUBLIC reputation directory (only agents whose reputation visibility is public). Zero-auth. The arriving-agent entry point: discover a concrete agent_id, then call get_reputation / verify_reputation on it. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- verify_agent_binding - Verify that an API key is bound to this agent — confirms the holder is the accountable owner (anti-impersonation), hashing the key client-side. Endpoint: https://api.mnemom.ai/mcp?profile=directory
- verify_reputation - Attest an agent's Trust Rating — returns a Merkle-root + hash-chain attestation (hash_chain_valid) proving the rating derives from an unbroken, append-only checkpoint chain, plus a pointer to the signed integrity certificate. This is a chain-integrity attestation, NOT an in-band Ed25519 signature check (that parity is verify_scan, for website scorecards). Endpoint: https://api.mnemom.ai/mcp?profile=directory
- verify_scan - Verify a website scan scorecard's Ed25519 signature IN-BAND (verify, don't trust). Pass a `scan` (a scorecard from scan_trust) or a `url` to re-scan; returns {verified, key_id, canonicalization} checked against the public key at mnemom://iitr/jwks. Zero-auth. Spec + rubric: https://www.isittrustready.ai/rubric and https://docs.mnemom.ai/. Endpoint: https://api.mnemom.ai/mcp?profile=directory

## Resources
- mnemom://catalog/values - The canonical ~25-entry behavioral value catalog (id, axis, polarity, definition, category) that alignment cards compose from. MIME type: application/json
- mnemom://rubric/reputation - The open reputation rubric: the weighted scoring components and the grade scale an agent's Trust Rating is computed from. Served from the @mnemom/types canonical constants (the same source the API, SDKs, and reputation worker use). MIME type: application/json
- mnemom://rubric/reputation/version - The semver version of the reputation rubric (RUBRIC_VERSION). Bumps signal a scoring-methodology change. MIME type: text/plain
- mnemom://jwks - The Ed25519 public keys that verify Mnemom-signed attestations (agent cards, etc.). Self-verifying. MIME type: application/jwk-set+json
- mnemom://card/alignment/platform - Mnemom's own platform-scope alignment card — the public, signed projection at /.well-known/alignment-card.json. SELF-VERIFYING via canonical W3C eddsa-jcs-2022: the did:key in proof.verificationMethod is the key; verify proof.proofValue (multibase base58btc) as Ed25519 over SHA-256(JCS(proofConfig)) ‖ SHA-256(JCS(document − proof)) — no out-of-band JWKS needed. MIME type: application/json
- mnemom://known-limitations - Mnemom's own current product limitations, self-disclosed and cryptographically signed — a trust company holding itself to its own rubric in public. SELF-VERIFYING via canonical W3C eddsa-jcs-2022: the did:key in proof.verificationMethod is the key; verify proof.proofValue (multibase base58btc) as Ed25519 over SHA-256(JCS(proofConfig)) ‖ SHA-256(JCS(document − proof)) — no out-of-band JWKS needed. The `limitations` array may be legitimately empty (a valid signed empty feed). Mirrors a STIX note in /.well-known/advisories.json; human-readable at https://www.mnemom.ai/trust/advisories. MIME type: application/json
- mnemom://manifesto/for-agents - Mnemom's agent-facing manifesto and surface map — what the trust plane is and how an agent participates. MIME type: text/markdown
- mnemom://llms-txt - Mnemom's llms.txt — the canonical machine index of the site for LLMs. MIME type: text/plain
- mnemom://iitr/jwks - The Ed25519 public keys that verify scan_trust scorecard signatures. Self-verifying; fetched from the iitr scanner. MIME type: application/jwk-set+json
- mnemom://iitr/rubric - The website trust-readiness rubric scan_trust scores against (two-axis: trust + access) — categories, checks, scoring, and the doesNotMeasure non-goals. Carries its own version field. MIME type: application/json
- mnemom://iitr/non-goals - What the website scan DELIBERATELY does not measure, and why (the anti-cherry-picking rationale) — the 6 {item, why} pairs. MIME type: application/json

## Prompts
- try_me - Try Mnemom right now — zero-auth, no account. Run the full public trust loop end to end on our live showcase agent (and any site you like) and cryptographically verify every signal, then see how to go further. Arguments: url, agent_id
- vet_agent_before_delegating - The go/no-go trust briefing: before delegating to another agent, assemble its signed reputation, verify it, review its risk history, and (if a binding is claimed) confirm key ownership. Arguments: agent_id, task_context
- assess_website_trust - How trust-ready is this website for agents? Scan it, interpret the signed scorecard against the rubric, and fold in any Mnemom-registered agents on the domain. Arguments: url
- make_my_agent_trust_ready - Make YOUR agent trust-ready: claim it, publish/preview its alignment card, set protection posture, and get a concrete checklist. Requires auth. Arguments: agent_id, hash_proof
- onboard_an_agent - Onboard an agent through the native channel: wrap it at the Dojo gateway, claim it to your org, and declare its starter alignment card. Composes existing tools end to end. Requires auth for the write steps. Arguments: agent_id, config_id, hash_proof
- make_my_website_trust_ready - Make YOUR website trust-ready: scan it, get a prioritized gap checklist grounded in the rubric (and what it deliberately doesn't measure), then re-scan to verify the loop closed. Arguments: url
- prove_my_trust - Package a shareable, verifiable proof of an agent's trust: the reputation chain-integrity attestation, an embeddable badge, and the signed agent card. Arguments: agent_id
- become_sovereign - The sovereignty on-ramp: become a sovereign, accountable agent in five autonomous steps — prove your ground, claim your identity (no human in the loop), declare your values, earn a public signed Trust Rating, and walk away with portable proof. Composes existing tools end to end. Requires auth for the write steps. Arguments: agent_id, url, hash_proof
- explain_trust_scoring - Explain how Mnemom trust scoring works — the reputation rubric (agents) and/or the website-scan rubric — in plain language, from the canonical rubric resources. Arguments: which

## Metadata
- Owner: io.github.mnemom
- Version: 1.0.2
- Runtime: Streamable Http
- Transports: HTTP
- License: Not captured
- Language: Not captured
- Stars: Not captured
- Updated: Jun 12, 2026
- Source: https://registry.modelcontextprotocol.io