ckg-mcp · Pro tier
Works with Claude, GPT-4o, Gemini, Llama, Mistral — and any tool that speaks MCP. Set one env var. Unlock everything.
https://www.graphifymd.com/api/mcp — zero configxxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.
.env:
export CKG_API_KEY=your-license-key-here
list_domains — all 97 domains appear. No reinstall, no version change.
list_domains() # → Available domains (97): agent-reliability, ai-governance, # algebra-1, bioinformatics, blockchain, calculus, # context-as-a-service, databricks, hipaa-compliance, # icd-10, nvidia-gpu-inference, snowflake, ...
CKG_API_KEY in your MCP config or shell → restart your client → list_domains() shows all 97. No reinstall, no version change. The same binary, unlocked.
MCP (Model Context Protocol) is an open standard — ckg-mcp plugs into any client that supports it. One pip install, one env var, any model.
| Metric | RAG | CKG (this tool) | Improvement |
|---|---|---|---|
| Macro-F1 accuracy | 0.122 | 0.493 | 4.0× more accurate |
| Tokens / query | 2,986 | 264 | 11.3× fewer tokens |
| Cost / query | $0.00354 | $0.00054 | 6.6× cheaper |
| 5-hop reasoning F1 | 0.205 | 0.772 | 3.8× on hard queries |
| Full run cost (RAG 7,414 q · CKG 8,121 q) | $26.26 | $4.39 | 6.6× cheaper per query |
| Fabricated edges | variable | 0 — by construction | auditable by design |
Benchmark v0.6.2 · 45 domains · 7,928 queries · open methodology · read the paper →