MNE-MCP
Brings the full MNE-Python neurophysiology stack into your MCP client with 38 structured tools spanning the standard pipeline: load EEG/MEG/fNIRS recordings, apply filters and re-referencing, fit ICA for artifact removal, epoch around events, compute ERPs and time-frequency representations, run source localization and connectivity analysis. Maintains a persistent session so you don't reload large raw files between preprocessing steps, saves every plot as PNG for the assistant to interpret, and includes an mne_run_code escape hatch when you need the entire MNE API. Ships with a suite of agent skills (mne-analyst, mne-methodology-critic, plus per-category helpers) and an interactive config wizard to set your line frequency, montage, filter bands, rejection thresholds, and ICA defaults once. Works across Windows, macOS, and Linux with any MCP client.
- GitHub stars
- 1
- tools
- 41
- transport
- STDIO
- last updated
- Jun 5, 2026
Tools
41 toolsPublic tool metadata for what this MCP can expose to an agent.
mne_check_statusCheck capabilities (in Status & Session section).
Check capabilities (in Status & Session section).
No parameter schema in public metadata yet.
mne_session_info
No parameter schema in public metadata yet.
mne_describe
No parameter schema in public metadata yet.
mne_get_info
No parameter schema in public metadata yet.
mne_reset_session
No parameter schema in public metadata yet.
mne_run_code
No parameter schema in public metadata yet.
mne_get_config
No parameter schema in public metadata yet.
mne_list_files
No parameter schema in public metadata yet.
mne_load_raw
No parameter schema in public metadata yet.
mne_filter
No parameter schema in public metadata yet.
mne_resample
No parameter schema in public metadata yet.
mne_crop
No parameter schema in public metadata yet.
mne_set_montage
No parameter schema in public metadata yet.
mne_set_reference
No parameter schema in public metadata yet.
mne_mark_bad_channels
No parameter schema in public metadata yet.
mne_interpolate_bads
No parameter schema in public metadata yet.
mne_plot_psd
No parameter schema in public metadata yet.
mne_plot_raw
No parameter schema in public metadata yet.
mne_plot_sensors
No parameter schema in public metadata yet.
mne_fit_ica
No parameter schema in public metadata yet.
mne_plot_ica_components
No parameter schema in public metadata yet.
mne_plot_ica_sources
No parameter schema in public metadata yet.
mne_apply_ica
No parameter schema in public metadata yet.
mne_find_events
No parameter schema in public metadata yet.
mne_events_from_annotations
No parameter schema in public metadata yet.
mne_make_epochs
No parameter schema in public metadata yet.
mne_plot_epochs_image
No parameter schema in public metadata yet.
mne_average_evoked
No parameter schema in public metadata yet.
mne_plot_evoked
No parameter schema in public metadata yet.
mne_plot_topomap
No parameter schema in public metadata yet.
mne_compute_tfrMorlet/multitaper, custom cycles, ITC, trial power, baseline
Morlet/multitaper, custom cycles, ITC, trial power, baseline
No parameter schema in public metadata yet.
mne_tfr_morlet
No parameter schema in public metadata yet.
mne_decodeMVPA
MVPA
No parameter schema in public metadata yet.
mne_connectivity
No parameter schema in public metadata yet.
mne_compute_connectivitybands, pairs, estimators
bands, pairs, estimators
No parameter schema in public metadata yet.
mne_compute_noise_cov
No parameter schema in public metadata yet.
mne_make_forward
No parameter schema in public metadata yet.
mne_apply_inverse
No parameter schema in public metadata yet.
mne_plot_source_estimate
No parameter schema in public metadata yet.
mne_decoding_group_testparticipant-level max-T or cluster-corrected inference
participant-level max-T or cluster-corrected inference
No parameter schema in public metadata yet.
mne_save
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
MNE-MCP
English | 简体中文
A Model Context Protocol (MCP) server that gives AI assistants direct, conversational access to MNE-Python for analyzing human neurophysiology data — EEG, MEG, sEEG, ECoG, and fNIRS.
Describe your analysis in plain language — MNE-MCP loads your recording, runs the MNE pipeline (filtering, ICA, epoching, ERP/ERF averaging, time-frequency, source-level work via code), saves the figures, and explains the results.
Details
- Repository
- github.com/exekiel179/mne-mcp
- Package
mne-mcp- Transport
- STDIO
- Registry
- MCP registry, active