The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, specifically pull request descriptions and source code files.
Ingestion points: The skill reads external content through the Read, Grep, and Glob tools during its core workflow (Workflow Step 1).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its own instructions and potentially malicious instructions embedded in the code being reviewed.
Capability inventory: The skill's capabilities are limited to reading files and reporting findings. It does not have access to network tools, shell execution, or privilege escalation commands.
Sanitization: The skill does not perform sanitization of the input data before processing it for review.
[SAFE]: The skill follows security best practices by including checks for OWASP Top 10 vulnerabilities, hardcoded secrets, and SQL injection in its review criteria. The references to external work (Jesse Vincent's superpowers) are properly attributed and legitimate resources for code review best practices.

code-reviewer