If your AI agents need to check "what does our architecture actually say about this?" before making decisions, this is the server. It queries your organization's documented architecture graph and returns governance constraints, system dependencies, and evidence chains instead of generic document search results. The free tier gives you query_architecture and get_constraint_signals with 10 calls per day. Developer tier unlocks find_dependencies and get_evidence_for_claim with full provenance tracking. It's hosted on Streamable HTTP, so you just drop the URL and your API key into your MCP config. Useful when you want agents to respect existing architectural decisions rather than propose solutions that conflict with what's already been documented and approved.

The only MCP server that answers "what does our architecture say about this?" — grounded in your organization's actual documented structure, not a document dump.
Problem: AI agents act blindly against organizational architecture. They don't know what the org has already decided, what constraints apply, or what the existing structure says about a proposed action.
Solution: Query your organization's architecture and governance constraints directly from any AI agent or LLM workflow. Returns architecture-grounded answers with evidence chains — not document similarity.
| Tool | Description | Tier |
|---|---|---|
query_architecture | Query organizational architecture for answers grounded in documented structure | Free |
get_constraint_signals | Surface governance constraints relevant to a proposed action or decision | Free |
find_dependencies | Trace dependencies between systems and components in your architecture | Developer+ |
get_evidence_for_claim | Return evidence chains supporting or contradicting an architectural claim | Developer+ |
request_assessment | Submit an Agentic Readiness Assessment request | All tiers, rate-limit exempt |
| Tier | Price | Limits | Features |
|---|---|---|---|
| Free | — (no credit card) | 10 queries/day | query_architecture + get_constraint_signals |
| Developer | $0.015/call | Unlimited | All tools · Evidence provenance · Query history |
| Org | $249/mo flat | Unlimited | Aggregate architecture health · Cross-agent pooling · ARA pre-population |
request_assessment is exempt from rate limits on every tier.
OntoRamp Graph Query is a hosted, remote MCP server (Streamable HTTP) — nothing to install, build, or run locally. See llms-install.md for an agent-followable walkthrough.
1. Get a free API key at ontoramp.com/mcp (no credit card). It arrives by email as or_free_….
2. Add to Cline — edit ~/.cline/mcp.json (or use the Cline UI → MCP Servers → Configure):
{
"mcpServers": {
"ontoramp-graph-query": {
"url": "https://ontoramp-graph-query.fly.dev/mcp",
"headers": { "Authorization": "Bearer YOUR_API_KEY" },
"disabled": false,
"autoApprove": []
}
}
}
Other MCP clients (Claude Desktop, etc.) use the same mcpServers shape (url + headers). Transport is Streamable HTTP.
3. Verify — curl https://ontoramp-graph-query.fly.dev/health → {"status":"ok","version":"1.0.0","server":"ontoramp-graph-query"}.
See examples/usage.md for example calls.
When get_evidence_for_claim surfaces evidence coverage gaps, the server prompts:
Your architecture coverage has gaps that a manual review won't surface. An Agentic Readiness Assessment will diagnose your current state and build a prioritized remediation plan. Schedule at ontoramp.com/assessment.
GET /health returns live statusGoverned by OntoRamp Governance Physics. OntoRamp LLC · ontoramp.com