Dependency Map Builder

Name/description (dependency mapping, critical path,催办建议) align with the included resources and the provided script. The script implements structured_brief, directory/csv/pattern audits and templates that match the skill's stated uses; requiring python3 only is proportionate.

SKILL.md restricts behavior to read-only, audit, and review drafts and explicitly disallows direct system modifications. The runtime instructions tell the agent to run the bundled script (python3 scripts/run.py) or produce output from local templates if execution isn't possible. The script will read files and directories supplied as --input and can traverse many text file types (.md, .py, .js, .sh, .json, .csv, etc.) to build directory or pattern audits — this is coherent for a 'directory audit' feature but means outputs could include sensitive file contents if a user points the tool at broad or system directories. The skill itself does not execute remote commands, does not spawn shells beyond normal interpreter usage, and avoids curl|bash patterns.

No install spec — instruction-only with an included Python script. No downloads, package manager installs, or archived extracts are present. Requiring only python3 and standard library is low risk and proportionate.

The skill declares no required environment variables, no credentials, and no config paths. The script runs purely on local inputs and resources; that matches the stated purpose.

always is false and the skill is user-invocable. The skill does not request permanent presence or modify other skills or system-wide agent settings. It can write an output file if --output is provided, which is expected behavior.