The Agent Skills Directory

[SAFE]: The skill is entirely instructional and designed for code review tasks. It guides the agent to read local repository data and reference files contained within the skill's own directory. No security issues such as prompt injection, malicious command execution, or unauthorized network activity were found.- [COMMAND_EXECUTION]: The skill uses the git diff command to identify changes in the local repository for review. This is a legitimate and necessary operation for the skill's stated purpose.- [PROMPT_INJECTION]: The skill ingests untrusted data in the form of code changes and source files for analysis. While this creates a surface for indirect prompt injection (Category 8), the skill's read-only nature and lack of external network or write capabilities mitigate the risk. Ingestion is performed via git diff and file reading tools without sanitization or explicit boundary markers.

convention-code-review