The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several git and GitHub CLI (gh) commands to manage the repository state.
Evidence includes: git status, git diff, git push, and gh pr create in SKILL.md.
[SAFE]: The skill implements safety checks to prevent common data exposure risks.
Evidence: Explicitly instructs the agent to "Never stage secrets (.env, credentials, etc.)" in the commit process.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it analyzes and summarizes untrusted data (git logs and diffs) to generate PR titles and descriptions.
Ingestion points: Reads git history via git log <base>...HEAD and changes via git diff <base>...HEAD as described in SKILL.md.
Boundary markers: Absent. The skill does not define specific delimiters or instructions for the agent to ignore instructions embedded within the code diffs it summarizes.
Capability inventory: Uses git for repository management and gh for network-enabled operations (opening pull requests).
Sanitization: Absent. There is no logic provided to sanitize or escape the content of the diffs before they are processed by the agent's summarization logic.

wrap-up