The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/draft_release_notes.py executes the git binary to retrieve repository information.
Evidence: The run_git function uses subprocess.run to execute commands starting with git (e.g., git log).
Context: This is a legitimate requirement for the skill's stated purpose of release management. The implementation uses a list of arguments rather than a shell string, which follows security best practices for preventing command injection.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted commit messages from the repository history and includes them in its output.
Ingestion points: scripts/draft_release_notes.py ingests commit subjects and author names from the output of git log.
Boundary markers: The script does not utilize delimiters or specific instructions to the agent to ignore any potential commands embedded within the commit messages.
Capability inventory: The skill has the capability to execute local commands via subprocess.run and write files to the disk via the --out parameter.
Sanitization: There is no evidence of sanitization or filtering of the commit content before it is rendered into the release notes markdown template.

github-release-management