The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local system commands git to analyze the repository state and gh pr create to submit pull requests to GitHub. These actions are aligned with the skill's stated purpose.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it incorporates untrusted data from git commit messages and file diffs into its analysis. Ingestion points: Git commit messages and file diffs (Step 1). Boundary markers: No explicit markers or instructions are provided to the agent to distinguish between git data and control instructions. Capability inventory: The skill possesses the ability to create pull requests (gh pr create), which is a network-based write operation. Sanitization: The skill does not perform any sanitization or validation on the content retrieved from the git repository before processing it.

create-pr