The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.Popen and subprocess.run to execute git log commands to analyze repository history. The repository path is provided via the --repo argument, which defaults to the current directory. This execution is confined to Git operations and does not use shell=True, which minimizes risks.
[EXTERNAL_DOWNLOADS]: The skill identifies the networkx library as a requirement for graph analysis and community detection. This is a well-known and trusted package in the Python ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Git logs (e.g., author names, emails, and commit metadata). While this data is used to generate reports that an agent might interpret, the risk is mitigated by structured parsing into JSON and CSV formats.
Ingestion points: scripts/build_ownership_map.py and scripts/community_maintainers.py via stdout from git log.
Boundary markers: None identified.
Capability inventory: Local file writing (summary.json, CSV outputs) and subprocess execution (git).
Sanitization: Extracted metadata is validated (e.g., date parsing) and formatted into standard CSV and JSON schemas.

security-ownership-map