The Agent Skills Directory

[PROMPT_INJECTION]: The skill is subject to indirect prompt injection as it ingests and processes untrusted data from the git repository's history.
Ingestion points: Data enters the agent's context through git log, git show, and gh pr view commands defined in references/commands.md.
Boundary markers: There are no explicit markers or instructions in the workflow to differentiate between the agent's core instructions and the text retrieved from commit messages or PR descriptions.
Capability inventory: The skill utilizes the Bash tool (restricted to git commands), Read, Grep, Glob, and the Agent tool as specified in the allowed-tools of SKILL.md.
Sanitization: The skill does not perform sanitization, filtering, or validation on the content retrieved from version control before processing it.

git-investigate