Rubric Protocol MCP server
{
"mcpServers": {
"0xsims-rubric-protocol": {
"command": "npx",
"args": [
"-y",
"@rubric-protocol/mcp-server@latest"
],
"env": {
"RUBRIC_API_KEY": "<rubric_api_key>"
}
}
}
}{
"mcpServers": {
"0xsims-rubric-protocol": {
"command": "npx",
"args": [
"-y",
"@rubric-protocol/mcp-server@latest"
],
"env": {
"RUBRIC_API_KEY": "<rubric_api_key>"
}
}
}
}{
"mcpServers": {
"0xsims-rubric-protocol": {
"command": "npx",
"args": [
"-y",
"@rubric-protocol/mcp-server@latest"
],
"env": {
"RUBRIC_API_KEY": "<rubric_api_key>"
}
}
}
}{
"servers": {
"0xsims-rubric-protocol": {
"command": "npx",
"args": [
"-y",
"@rubric-protocol/mcp-server@latest"
],
"env": {
"RUBRIC_API_KEY": "${input:rubric_api_key}"
}
}
},
"inputs": [
{
"id": "rubric_api_key",
"type": "promptString",
"description": "Rubric api key",
"password": true
}
]
}{
"mcpServers": {
"0xsims-rubric-protocol": {
"command": "npx",
"args": [
"-y",
"@rubric-protocol/mcp-server@latest"
],
"env": {
"RUBRIC_API_KEY": "<rubric_api_key>"
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @rubric-protocol/mcp-serverversion 2.0.2
What it does
Rubric Protocol provides AI compliance attestation and verification tools for regulatory frameworks including the EU AI Act, SR 11-7, HIPAA, and NIST AI RMF. It generates cryptographically signed and blockchain-anchored proofs of AI decisions for compliance documentation and auditing.
- Generate and verify ML-DSA-65 signed attestations of AI model outputs for regulatory filing and audit trails.
- Export compliance evidence in EU AI Act Annex IV, OSCAL, C2PA, and W3C Verifiable Credential formats.
- Register AI agents and models, detect applicable frameworks, and assess regulatory jurisdiction requirements.
Tools this server exposes
| Tool | Does |
|---|---|
| attest | Create a cryptographically signed attestation of an AI decision. |
| verify | Verify a signed attestation and its blockchain anchor. |
| get_proof | Retrieve the proof record for an attestation. |
| register_agent | Register an AI agent in the compliance registry. |
| status | Check the operational status of the Rubric service. |
| framework_detect | Identify applicable regulatory frameworks for a given AI use case. |
| cost_estimate | Estimate costs for compliance attestation operations. |
| bundle_query | Query bundled attestations and proofs. |
| attest_batch | Create multiple attestations in a single batch operation. |
| verify_chain | Verify a chain of attestations. |
| verify_batch | Verify multiple attestations in batch. |
| annex4_generate | Generate EU AI Act Annex IV compliance documentation. |
| c2pa_attest | Create a C2PA manifest for an AI output. |
| credential_issue | Issue a W3C Verifiable Credential for compliance evidence. |
| nist_rmf_certify | Certify AI system compliance against NIST AI Risk Management Framework. |
| gpai_register | Register a general-purpose AI model for regulatory tracking. |
| jurisdiction_assess | Assess regulatory jurisdiction requirements for an AI application. |
| incident_create | Log and attest an AI-related incident. |
| agent_add | Add an AI agent to the registry. |
| model_register | Register an AI model in the compliance registry. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.
This server has not seen a commit in 2 mo. It may still work, but nobody is currently fixing it — check the alternatives before you build on it.