The Agent Skills Directory

[PROMPT_INJECTION]: Potential for indirect prompt injection through the analysis of external code provided in Pull Requests.
Ingestion points: The skill reads git diffs and file contents from the checked-out branch using git diff and file system tools (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the analyzed code.
Capability inventory: The skill uses git, Read, Grep, and Glob to inspect the filesystem and project history.
Sanitization: No sanitization or filtering is performed on the ingested code changes before they are processed by the agent.

pr-review