openinterp / mcp
Bring-your-own-agent infrastructure for mechanistic interpretability research. Works with Claude Code, Cursor, Cline, OpenHands, Aider — anything that speaks MCP. We never see your model, your data, or your keys.
The model runs in your Colab / vast.ai / runpod session. A FastAPI backend exposes 8 typed endpoints. An ngrok tunnel makes them reachable from your laptop. The MCP server on your laptop is stateless — it just routes tool calls.
USER'S LAPTOP USER'S COMPUTE (Colab / vast.ai / runpod)
├── Claude Code / Cursor / Cline ├── Colab Secrets (HF, OpenAI, Anthropic, ngrok)
├── openinterp-mcp (stateless tool router) ├── FastAPI with 8 typed endpoints
└── ~/.openinterp/sessions.json ├── HF model + tokenizer
(URLs only, no secrets) ├── Probe registry
└── ngrok HTTPS tunnel
←──── HTTPS (ngrok URL) ────────────────────┘
DOES NOT EXIST: hosted inference, key custody, telemetry, query database.colab_attachAttach to a running Colab session via its ngrok URL.
colab_statusHealth check, loaded model, probes in memory.
list_probesProbes currently in the backend registry.
capture_actsForward pass with hooks. Returns capture_id.
probe_evalApply a probe to a stored capture. AUROC if labels.
steerInject direction*α. Control-token-normalized Δrel returned.
sae_lookupDecompose an activation into top-K SAE features.
causality_protocolThree mandatory checks → verdict in 5 classes.
causality_protocol by default and emit a 5-class verdict.Apache-2.0. PyPI · GitHub · Quick start docs