The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git diffs, commit messages, and external project tracking systems. Malicious instructions embedded in code comments, pull request descriptions, or ticket comments could potentially influence the agent's logic or behavior during the review process. * Ingestion points: Data is fetched via git diff, commit logs, and external tracker bodies/comments as specified in the Process section of SKILL.md. * Boundary markers: The skill definition does not specify any delimiters or explicit instructions for the agent to ignore or isolate instructions found within the processed diff or tracker data. * Capability inventory: The skill possesses file read/write access (e.g., writing to .agents/projects/<project>/review.md) and utilizes command-line tools like git. * Sanitization: No specific sanitization, filtering, or validation of the ingested external content is defined in the instructions.

code-review