glassbox-mcp
MCP server for Glassbox — mechanistic interpretability + EU AI Act Annex IV compliance tools, exposed via the Model Context Protocol.
Connect Claude (or any MCP-compatible client) to Glassbox with a single pip install. No cloning, no manual server setup.
Install
pip install glassbox-mcp
Connect to Claude Desktop
Add to ~/.claude/claude_desktop_config.json (macOS/Linux) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"glassbox": {
"command": "glassbox-mcp"
}
}
}
Restart Claude Desktop. The Glassbox tools appear immediately.
Tools
| Tool | What it does |
|---|---|
glassbox_circuit_discovery |
Attribution patching — find which attention heads causally drive a prediction |
glassbox_faithfulness_metrics |
Compute sufficiency, comprehensiveness, F1, and EU AI Act explainability grade (A–D) |
glassbox_compliance_report |
Generate a full EU AI Act Annex IV evidence package (all 9 sections) |
glassbox_attention_patterns |
Get attention weight heatmap for any layer/head |
glassbox_logit_lens |
Layer-by-layer residual stream projection — see how predictions build up |
Supported Models
gpt2,gpt2-medium,gpt2-largeEleutherAI/pythia-70m,pythia-160m,pythia-410m
Example usage in Claude
Once connected, ask Claude things like:
"Use Glassbox to analyse which attention heads drive GPT-2's prediction for 'The Eiffel Tower is in' → ' Paris'"
"Generate an EU AI Act Annex IV compliance report for GPT-2 with provider name Acme Corp"
"Show me the logit lens for GPT-2 on this sentence"
Requirements
- Python ≥ 3.10
- ~3 GB disk (PyTorch + TransformerLens model weights download on first use)
Version
4.2.4 — tracks glassbox-mech-interp v4.2.4
- Paper: arXiv 2603.09988
- PyPI (core library): glassbox-mech-interp
- Website: project-gu05p.vercel.app
- GitHub: designer-coderajay/Glassbox-AI-2.0
Metadata
Release files for glassbox-mcp 4.2.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| glassbox_mcp-4.2.4.tar.gz | 11.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| glassbox_mcp-4.2.4-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 23.0 kB
Release files / glassbox_mcp-4.2.4.tar.gz
| Download URL | glassbox_mcp-4.2.4.tar.gz |
|---|---|
| Size | 11.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
6f653e51527ade7865ad39086af39d8956ca20ec762c62762d4241076dde7b49
|
|
BLAKE2b-256 checksum How to use checksums |
b202e85fa806e441a000f854e1246b585414e74e3a8f3f5a524502d538c57e2b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Release files / glassbox_mcp-4.2.4-py3-none-any.whl
| Download URL | glassbox_mcp-4.2.4-py3-none-any.whl |
|---|---|
| Size | 11.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e85b76befbdd21ba0fbbf2115b03d5e6c3fe7c8df31c8d56149f8280a7d0f3d5
|
|
BLAKE2b-256 checksum How to use checksums |
31b03aea9ed673835c128b561af2347322cae939a4fc01e24cf84c19452ec875
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|