MCP API Reference

OntoRamp exposes eleven tools across three MCP plugins: Knowledge Graph (governance and compliance search), Governance Evaluator (maturity gap analysis and document validation), and Projection Engine (governance assessment and domain-level findings). All tools are available via the MCP Streamable HTTP transport; legacy SSE-named aliases remain supported. This page documents every tool's parameters and response shape. Search relevance is returned as a coarse bucket — high, medium, or low — not a raw score.

Authentication

Every request requires an API key. Pass it in the Authorization header as a Bearer token (recommended), or as a query parameter on the legacy connection URL.

# Streamable HTTP endpoint (recommended)
POST https://mcp.ontoramp.com/{plugin}/mcp
Authorization: Bearer YOUR_API_KEY

# Legacy alias (query-parameter auth, still supported)
POST https://mcp.ontoramp.com/{plugin}/sse?api_key=YOUR_API_KEY

Replace {plugin} with one of: knowledge-graph, governance-evaluator, projection-engine.

{
  "results": [
    {
      "chunk_id": "uuid",
      "content_text": "...",
      "relevance": "high",
      "domain": "governance",
      "title": "...",
      "source_kind": "Governance Reference"
    }
  ],
  "total": 10,
  "query": "..."
}

{
  "results": [
    {
      "chunk_id": "uuid",
      "content_text": "...",
      "relevance": "high",
      "domain": "governance",
      "title": "...",
      "source_kind": "Governance Reference"
    }
  ],
  "total": 10,
  "query": "..."
}

{
  "results": [
    {
      "chunk_id": "uuid",
      "content_text": "...",
      "relevance": "high",
      "domain": "governance",
      "title": "...",
      "source_kind": "Governance Reference"
    }
  ],
  "total": 10,
  "query": "..."
}

{
  "results": [
    {
      "entity_title": "Data Governance Board",
      "entity_type": "role",
      "definition_text": "...",
      "synonyms": ["..."],
      "domain": "governance",
      "relevance": "high",
      "chunk_ids": ["uuid", "uuid"]
    }
  ],
  "total": 10,
  "query": "..."
}

{
  "edges": [
    {
      "edge_id": "uuid",
      "source_document_uid": "uuid",
      "target_document_uid": "uuid",
      "source_entity_uid": null,
      "target_entity_uid": null,
      "edge_family": "depends on",
      "relevance": "high",
      "source_domain": "governance",
      "target_domain": "security",
      "hop_distance": 1
    }
  ],
  "total": 12,
  "document_uid": "uuid",
  "depth": 1
}

{
  "domain": "governance",
  "total_entities": 128,
  "total_chunks": 512,
  "coverage_by_type": [
    {
      "entity_type": "governance_concept",
      "count": 42,
      "percentage": 32.8
    }
  ],
  "gaps": ["..."],
  "top_entities": [
    { "title": "...", "type": "pattern" }
  ],
  "assessment_cta": "..."
}

{
  "findings": [
    {
      "rule": "...",
      "severity": "warning",
      "message": "...",
      "suggestion": "..."
    }
  ],
  "entity_coverage": {
    "matched_entities": [
      {
        "entity_title": "...",
        "entity_type": "governance_concept",
        "relevance": "high"
      }
    ],
    "total_matched": 11
  },
  "domain_analysis": {
    "detected_domains": [
      { "domain": "governance", "chunk_count": 8 }
    ],
    "expected_domain": "governance",
    "aligned": true
  },
  "summary": {
    "total_findings": 4,
    "errors": 0,
    "warnings": 2,
    "info": 2,
    "entity_coverage_score": 0.79
  }
}

{
  "topic": "...",
  "brief": "...",
  "sources": [
    {
      "chunk_id": "uuid",
      "content_preview": "...",
      "domain": "governance",
      "relevance": "high"
    }
  ],
  "entities": [
    {
      "entity_title": "...",
      "entity_type": "governance_concept",
      "definition_text": "..."
    }
  ],
  "domain_coverage": [
    { "domain": "governance", "source_count": 3 }
  ],
  "metadata": {
    "depth": "summary",
    "sources_used": 5,
    "entities_found": 3,
    "model": "..."
  }
}

{
  "runs": [
    {
      "run_id": "uuid",
      "status": "completed",
      "assessment": {
        "overall_score": 78.5,
        "posture_state": "...",
        "domain_scores": {
          "GOVERNANCE": 88.1,
          "PLATFORM": 76.4,
          "SECURITY": 79.3,
          "AI": 82.0,
          "DATA": 65.0
        }
      },
      "assessment_summary": {
        "artifacts_assessed": 512,
        "domain_coverage": { "governance": 128 }
      },
      "started_at": "2026-04-06T10:00:00Z",
      "completed_at": "2026-04-06T10:00:03Z",
      "source": "own",
      "is_bootstrap": false
    }
  ],
  "total": 1
}

{
  "packets": [
    {
      "packet_id": "uuid",
      "simulation_run_id": "uuid",
      "domain": "governance",
      "question": "...",
      "recommendation": "APPROVE",
      "confidence": 0.917,
      "rationale": "...",
      "governed_by": "...",
      "score_summary": {
        "overall_score": 91.7,
        "posture_state": "...",
        "domain_scores": { "GOVERNANCE": 88.1 }
      },
      "compliance_summary": {
        "checks_passed": 6,
        "checks_total": 6
      },
      "created_at": "2026-04-06T10:00:00Z",
      "source": "own",
      "is_bootstrap": false
    }
  ],
  "total": 6
}

{
  "run_id": "uuid",
  "status": "completed",
  "overall_score": 78.5,
  "posture_state": "...",
  "domain_scores": {
    "GOVERNANCE": 88.1,
    "PLATFORM": 76.4,
    "SECURITY": 79.3,
    "AI": 82.0,
    "DATA": 65.0
  },
  "packets_created": 6,
  "persisted": true,
  "started_at": "2026-04-06T10:00:00Z",
  "completed_at": "2026-04-06T10:00:03Z"
}

Rate limits

When a request is rate-limited, the rejection itself is the signal — an HTTP 429 response or an in-band RATE_LIMITED error in the tool result, depending on the plugin. Responses do not carry X-RateLimit-* headers; on a rejection, back off and retry after the reset.

Hitting the free-tier quota? Org tier is unlimited at a flat monthly rate — see plugin pricing.

Error codes