Rubric MCP server

Read from the package source without running it. The installed server may list more.

• attest

  Attest an AI decision. With RUBRIC_API_KEY: HCS-anchored on Hedera mainnet. Without: PQ-signed local Merkle leaf in ~/.rubric/local-bundles/.

• bundle_query

  Query attestation bundles by leaf type, agent, or time range. Requires Standard+ tier.

• check_counterparty

  Get an attested OFAC/sanctions + on-chain risk screening for a Base address before transacting with it. The verdict is Hedera-anchored and citable as proof of diligence. Paid x402 service ($0.01 USDC); returns the exact payment challenge so a wallet-capable agent can obtain the live signed result.

• cost_estimate

  Estimate monthly Rubric cost from expected decision volume. No key required.

• explain_rubric

  What Rubric is and why it matters: the difference between logs you control (testimony) and anchored records (evidence). Returns the trust model, how to verify independently, and pricing. No key required.

• framework_detect

  Auto-detect applicable regulatory frameworks (EU AI Act, SR 11-7, HIPAA, NIST AI RMF, etc.) from decision content. Works offline, no key required.

• get_proof

  Generate ZK Merkle inclusion proof (Noir/Barretenberg) for an attestation. Requires API key.

• notarize

  Notarize any JSON record: prove it existed, unaltered, at a point in time. Returns a PROOF FILE the caller must keep - the chain stores a commitment, never the record. Paid x402 ($0.005 USDC) or with an API key. Proves existence and integrity, NOT that the record is true.

• register_agent

  Register an agent and receive a free Rubric developer API key via email.

• relay_paywrite action

  Pay any x402 endpoint through Rubric and receive a signed, Hedera-anchored receipt of what was actually delivered: request hash, response hash, byte count, status, settlement reference. The provider is paid directly by you and Rubric never holds or submits your funds. Use this instead of calling an endpoint directly when you need provable evidence of what you received, not just the response. Free

• screen_name

  Screen a person or entity name against six sanctions and export-control lists: OFAC SDN and Consolidated (with alternate spellings), UN Security Council, UK OFSI, EU, and BIS Denied Persons. Fuzzy matching with the full rule, list versions, and per-hit edit distance disclosed in every response. Matches are CANDIDATES requiring human adjudication, not determinations. Paid x402 service ($0.01 USDC);

• status

  Federation health across US/SG/JP/CA/EU nodes + ZK node.

• verify

  Verify an attestation. Checks local store first, then federation. Merkle inclusion + HCS anchoring.

• verify_record

  Verify a notarization. Two independent checks: does the record hash match the proof file (offline), and is the attestation anchored on Hedera consensus (online). Always free, no key, no payment - a proof only the issuer can check is not a proof.

• Code scan4 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitylocal package, no credentials required12/15
• Maintenancerepository not readable: repo not found3/15
• Maintainer identityno repository or website to verify2/10

Safety score

65/100

higher than 25% of 4,461 AI and agents servers

Tools

14

the typical AI and agents server has 6

Tools that change things

7%

typical: 0%. Fewer write tools means less an agent can break

Downloads a week

57

more than 46% of 1,554 AI and agents servers with a package

Rubric: common questions

Is Rubric MCP server safe?

With care: it is graded C, so read the findings first (65/100). Read the Rubric safety report

How do I install Rubric?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Rubric need an API key?

Not as far as the registry entry and our scan can tell: no credentials are declared or required.

Is Rubric maintained?

The latest release is v2.5.0.

What can I use instead of Rubric?

Servers from other publishers that do the same job: Watch Skill MCP server, Mockserver MCP server and CHAP Coordinator MCP server.

How does Rubric compare with other AI and agents MCP servers?

safety score 65/100 (higher than 25% of 4,461 AI and agents servers); tools 14 (the typical AI and agents server has 6); tools that change things 7% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.