Rubric Mcp Server — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Rubric Mcp Server (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
AI compliance attestation for Claude Desktop, Cursor, Windsurf, and any MCP-compatible host.
Every AI decision PQ-signed, Merkle-chained, and anchored to Hedera mainnet. Built for EU AI Act Article 12, SR 11-7, HIPAA, and NIST AI RMF.
Add to your claude_desktop_config.json:
{ "mcpServers": { "rubric": { "command": "npx", "args": ["-y", "@rubric-protocol/mcp-server@latest"], "env": { "RUBRIC_API_KEY": "your-key-here" } } } }
Free local tier, no key required:
{ "mcpServers": { "rubric": { "command": "npx", "args": ["-y", "@rubric-protocol/mcp-server@latest"] } } }
Restart Claude Desktop. Rubric tools appear automatically.
47 tools across 8 modules. Default profile exposes the core 8; set RUBRIC_MCP_MODULES=all (or a comma list, e.g. core,compliance) for more.
core (default): attest, verify, get_proof, register_agent, status, framework_detect, cost_estimate, bundle_query attestation: attest_batch, attestation_status, attestation_get, pipeline_trace, bundle_get verification: verify_chain, verify_tree, verify_batch, zk_verify, zk_proof_get, ledger_lookup compliance: annex4_generate, annex4_status, c2pa_attest, c2pa_assertion, credential_issue, credential_get, compliance_query, compliance_report, filing_generate regulatory: gpai_register, gpai_downstream, nist_rmf_certify, nist_rmf_status, jurisdiction_map, jurisdiction_assess, jurisdiction_gap governance: incident_create, incident_attest, incident_resolve, human_review, adversarial_session_start, adversarial_session_conclude registry: agent_add, agent_get, model_register, model_get ops: usage_report, auditor_token_create
The only MCP tools whose outputs are ML-DSA-65 (FIPS 204) signed, Hedera-anchored, and exportable as EU AI Act Annex IV, OSCAL, C2PA, and W3C Verifiable Credential evidence.
Run register_agent directly inside Claude Desktop, no browser required. Or visit https://rubric-protocol.com
EU AI Act | SR 11-7 | NIST AI RMF | HIPAA | NYC Local Law 144 | ECOA Reg B
https://rubric-protocol.com https://www.npmjs.com/package/@rubric-protocol/mcp-server https://registry.modelcontextprotocol.io
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.