Ami Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Ami Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
[![Actions Status][actions-badge]][actions-link] [![Documentation Status][rtd-badge]][rtd-link]
[![PyPI version][pypi-version]][pypi-link] [![PyPI platforms][pypi-platforms]][pypi-link]
[![GitHub Discussion][github-discussions-badge]][github-discussions-link]
[![Coverage][coverage-badge]][coverage-link]
<!-- --8<-- [start:intro] -->
An MCP server that wraps ATLAS AMI (ATLAS Metadata Interface) and the PMG cross-section database, exposing them as tools for LLMs. Designed for ATLAS physicists who need to discover MC samples, look up cross-sections and filter efficiencies, and validate PMG hashtag classifications.
<!-- --8<-- [end:intro] -->
<!-- --8<-- [start:what-it-does] -->
ami-mcp lets Claude (or any MCP-compatible LLM) query ATLAS metadata directly:
(WeakBoson/Vjets/Baseline), by name pattern, or by arbitrary AMI query
dataset provenance, and AMI processing tag info
files (PMGxsecDB_mc16.txt, etc.)
against the PMG cross-section database
LLM using the ami://query-language resource as a guide
<!-- --8<-- [end:what-it-does] -->
<!-- --8<-- [start:installation] -->
pip install ami-mcpOr with pixi (recommended for ATLAS facilities):
pixi add ami-mcp<!-- --8<-- [end:installation] -->
<!-- --8<-- [start:requirements] -->
<3.12)voms-proxy-init -voms atlas)<!-- --8<-- [end:requirements] -->
<!-- --8<-- [start:quick-start] -->
voms-proxy-init -voms atlasWhen installed via pip (not pixi/conda-forge), also set X509_CERT_DIR. On CVMFS-based facilities (e.g. UChicago Analysis Facility, CERN lxplus):
export X509_CERT_DIR=/cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase/etc/grid-security-emi/certificatesWhen installed via pixi or conda-forge, ca-policy-lcg is included and sets X509_CERT_DIR automatically — no extra step needed.
ami-mcp serveThe server speaks MCP over stdio. Configure your MCP client to launch it.
With pixi (recommended — X509_CERT_DIR is set automatically):
{
"mcpServers": {
"ami": {
"type": "stdio",
"command": "pixi",
"args": ["run", "ami-mcp", "serve"],
"env": {
"ATLAS_PMGXSEC_PATH": "/cvmfs/atlas.cern.ch/repo/sw/database/GroupData/dev/PMGTools"
}
}
}
}With pip (must set X509_CERT_DIR manually):
{
"mcpServers": {
"ami": {
"command": "ami-mcp",
"args": ["serve"],
"env": {
"X509_CERT_DIR": "/cvmfs/atlas.cern.ch/repo/ATLASLocalRootBase/etc/grid-security-emi/certificates",
"ATLAS_PMGXSEC_PATH": "/cvmfs/atlas.cern.ch/repo/sw/database/GroupData/dev/PMGTools"
}
}
}
}Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"ami": {
"command": "ami-mcp",
"args": ["serve"],
"env": {
"X509_CERT_DIR": "/path/to/ca-certificates",
"ATLAS_PMGXSEC_PATH": "/path/to/PMGTools"
}
}
}
}<!-- --8<-- [end:quick-start] -->
<!-- --8<-- [start:tools] -->
| Tool | Description |
|---|---|
ami_execute | Execute any AMI command string (primary power tool) |
ami_get_dataset_info | Get metadata for a dataset (nFiles, nEvents, status, etc.) |
ami_get_dataset_prov | Get provenance chain with lineage summary; filter by data type |
ami_list_datasets | Search for datasets by name pattern (supports % wildcards) |
| Tool | Description |
|---|---|
ami_search_by_hashtags | Find datasets by hashtag combination (e.g. WeakBoson/Vjets/Baseline) |
ami_get_dataset_hashtags | Look up PMGL1–PMGL4 classification for a dataset |
| Tool | Description |
|---|---|
ami_get_physics_params | Get cross-section (nb→pb converted), filter efficiency, k-factor from AMI |
ami_get_ami_tag | Get AMI processing tag info; accepts tag chains like e8351_s3681_r13144 |
| Tool | Description |
|---|---|
ami_list_xsec_databases | List available PMGxsecDB\_\*.txt files |
ami_lookup_xsec | Look up DSID cross-section, filter eff, k-factor in xsec DB |
| Tool | Description |
|---|---|
ami_validate_sample | Check hashtag classification and compare metadata to xsec DB |
<!-- --8<-- [end:tools] -->
<!-- --8<-- [start:example-prompts] -->
Once configured, you can ask Claude things like:
<!-- --8<-- [end:example-prompts] -->
<!-- prettier-ignore-start --> [actions-badge]: https://github.com/kratsg/ami-mcp/actions/workflows/ci.yml/badge.svg [actions-link]: https://github.com/kratsg/ami-mcp/actions [github-discussions-badge]: https://img.shields.io/static/v1?label=Discussions&message=Ask&color=blue&logo=github [github-discussions-link]: https://github.com/kratsg/ami-mcp/discussions [pypi-link]: https://pypi.org/project/ami-mcp/ [pypi-platforms]: https://img.shields.io/pypi/pyversions/ami-mcp [pypi-version]: https://img.shields.io/pypi/v/ami-mcp [rtd-badge]: https://readthedocs.org/projects/ami-mcp/badge/?version=latest [rtd-link]: https://ami-mcp.readthedocs.io/en/latest/?badge=latest [coverage-badge]: https://codecov.io/github/kratsg/ami-mcp/branch/main/graph/badge.svg [coverage-link]: https://codecov.io/github/kratsg/ami-mcp
<!-- prettier-ignore-end -->
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.