The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool restricted to specific git commands including git log, git diff, and git show. These are used exclusively for read-only analysis of the repository history to identify changes, migrations, and dependency updates.
[PROMPT_INJECTION]: The skill has an attack surface for Indirect Prompt Injection (Category 8) because it ingests and processes untrusted data from the repository's git history.
Ingestion points: Git commit messages, diff content, and pull request metadata are read into the agent's context during Step 1 (Analyze Changes).
Boundary markers: The prompt does not define clear delimiters or "ignore embedded instructions" warnings for the content retrieved from git commands.
Capability inventory: The skill's capabilities are restricted to file system reading (Read, Glob, Grep) and git inspection; it lacks the ability to write files, execute arbitrary shell scripts, or perform network requests.
Sanitization: No explicit sanitization or filtering is performed on the text of commit messages or diffs before they are processed by the LLM.

deploy-checklist