agentleFS
Sign inStart free
MicrosoftKnown publishermicrosoft/healthcareai-examples/agentic/medimageinsight/skills/ground-adapter/SKILL.md

ground-adapter

Skill204 starsChanged 4 months agoMITView on GitHub

Ground CXR findings using the trained MI2 adapter. Returns per-finding positive/possible/negative calls for the agent to use when writing the report.

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not retry on success.

    line 16
  2. Very specific:

    Call the adapter: Call mcp_medimageinsig_adapter_classify with file_path = absolute path and head = the caller's value if provided …

    line 22

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ground-adapter

Calls the trained MI2 adapter and returns per-finding positive/possible/negative calls. Does not write a report — the agent produces the report using the write-report skill after grounding.

Budget

  • Tool calls: 1 — exactly one mcp_medimageinsig_adapter_classify call per image.
  • Do not retry on success.

How to use

  1. Resolve the image. If image_path is an http(s) URL, pass it through to the tool unchanged — do NOT download it, curl it, or turn it into a local file; the MCP server fetches URLs itself. Otherwise it is a local path: the MCP tool runs in a different working directory, so if it is not absolute, resolve it via realpath; if it can't be resolved, ask the user.
  2. Call the adapter. Call mcp_medimageinsig_adapter_classify with file_path = absolute path and head = the caller's value if provided (otherwise omit; defaults to mlp). One call per image.
  3. Read the results. The adapter returns {label: "positive" | "possible" | "negative"}, already bucketed using per-class operating points fit during training (Youden-J for rule-in, sens90 for rule-out). No further thresholding.
  4. Return the findings. Return the positive and possible labels. Omit negatives. Do not write a report. The agent decides what to do next.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free