grid-atlas-qc

cryo-EMMCP servercommunity

Scores grid squares and foil holes from an atlas montage, estimating ice thickness and contamination to produce a ranked acquisition queue for automated data collection.

tool call

score_atlas

license

MIT

runtime

python 3.11

stars

★ 0

// emskills.config.json

"skills": ["grid-atlas-qc"]

$ npx emskills add grid-atlas-qc

Open on GitHub ↗

Test runs are served by the registry's own sandbox, executed by a model grounded in this SKILL.md — a dry run of the procedure, not the vendor's production code.

Documentation

This skill exposes the call score_atlas, running on python 3.11. Install it in your agent with the config below, or exercise it from here — the test panel above sends a real request to the endpoint and prints every step.

1 — install in an agent host (Claude, Cursor, ChatGPT)

{
  "mcpServers": {
    "grid-atlas-qc": {
      "type": "http",
      "url": "https://emskills.org/api/public/mcp/grid-atlas-qc"
    }
  }
}

2 — run it locally through the emskills CLI

{
  "mcpServers": {
    "grid-atlas-qc": {
      "command": "npx",
      "args": [
        "-y",
        "emskills",
        "run",
        "grid-atlas-qc"
      ]
    }
  }
}

3 — call score_atlas directly with curl

curl -s https://emskills.org/api/public/mcp/grid-atlas-qc \
  -H 'content-type: application/json' \
  -H 'accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"score_atlas","arguments":{"input":"./dataset.mrc"}}}'

4 — example arguments for the test panel

{
  "input": "./dataset.mrc",
  "pixel_size_nm": 0.86
}

Related cryo-EM skills

cryo2d-classify

Reference-free 2D classification of particle micrographs with a motion-correction pre-pass.

particle-picker

Template-free particle picking tuned for low signal-to-noise micrographs