Ephys MCP server

ephys-mcp

<!-- mcp-name: io.github.happyc0der/ephys-mcp -->

An MCP server that lets an LLM analyse intracortical (spike-level) brain-computer-interface recordings: signal quality, spike detection, firing rates, and cursor-velocity decoding.

Existing BCI MCP servers target scalp EEG. This one targets the kind of data a high-channel-count implant produces, and defines a read-only adapter contract so a live device backend can be added when a vendor publishes an API.

Research and education software. Not a medical device. Not for clinical use.

Not affiliated with or endorsed by Neuralink Corp. or any other implant manufacturer.

Status

v0.5. The planned feature set is complete: local NWB files, local broadband WAV recordings, live Lab Streaming Layer streams, streaming from the DANDI Archive, a synthetic motor-cortex source with ground truth, spike detection, quality metrics, ridge and Kalman decoders, trial-aligned PSTHs, spike sorting, cross-session (FALCON-style) evaluation, latent-factor models (GPFA, PCA), probe geometry, and figures. 24 tools. Bug reports and feature requests go to GitHub Issues.

Install and run

Needs uv. No install step: uvx ephys-mcp fetches the package and starts the server on stdio.

Claude Code:

claude mcp add ephys -- uvx ephys-mcp

Claude Desktop (claudedesktopconfig.json):

{ "mcpServers": { "ephys": { "command": "uvx", "args": ["ephys-mcp"] } } }

From a checkout, use uv run ephys-mcp instead, or uv --directory /path/to/ephys-mcp run ephys-mcp in the configs above.

HTTP transport

For remote clients or hosted agents, serve streamable HTTP instead of stdio:

EPHYS_MCP_TOKEN='a-long-random-secret' uvx ephys-mcp --http --host 0.0.0.0 --port 8000

Every request must then carry Authorization: Bearer . The server refuses to bind to a non-loopback address without a token, and tokens must be at least 16 characters. Put TLS in front of it (a reverse proxy) before exposing it beyond a private network: the token travels in clear text otherwise. On loopback the token is optional, so ephys-mcp --http alone serves http://127.0.0.1:8000/mcp for local testing.

Then ask, for real data: "Find a small motor cortex dataset on DANDI, open it, and tell me how well hand velocity can be decoded." Or offline: "Open a synthetic session, check signal quality, fit a Kalman decoder and show me a decoded window."

Data sources

Dataset licence and citation come from the archive and are returned by opensession, so the model can attribute the data. Many datasets record only during trials; the server tracks those spans (recordedfraction) and leaves the gaps out of rates and decoding instead of reading them as silence.

WAV samples carry no physical unit, so amplitudes are reported as ADC counts unless you pass uvpercount; every amplitude result names its unit. Clips in a folder are separate recordings, so the server says that timing across those channels is not meaningful. Spike times from WAV are threshold crossings, not sorted units.

Shortened. The full README is on GitHub.