The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/repo-doctor.py executes git commands using subprocess.run to audit repository state, including git rev-list and git log. While restricted to local git operations, this involves shell command execution.
[INDIRECT_PROMPT_INJECTION]: The skill audits external repositories, reading file names, contents, and commit messages. This creates a surface where malicious content within an audited repository could influence the agent's subsequent remediation steps.
Ingestion points: scripts/repo-doctor.py reads file contents and git logs from the target repository.
Boundary markers: The script does not implement explicit boundary markers when reporting findings to the agent.
Capability inventory: The skill can read files and execute git via subprocess.
Sanitization: The script uses UTF-8 error handling (errors='replace') but does not sanitize content against prompt injection patterns.

repo-doctor