The Agent Skills Directory

[SAFE]: The skill's primary purpose is to assist developers with security reviews by analyzing git diffs and searching for common vulnerability patterns (OWASP Top 10, etc.). Its behavior is consistent with its stated goal.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to ingest and process untrusted data (source code from the project being audited).
Ingestion points: The skill reads all files identified via git diff and CLAUDE.md to perform the audit.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its own instructions and potential malicious instructions embedded within the files it is auditing.
Capability inventory: The skill has access to shell commands (specifically git operations) and the WebFetch tool.
Sanitization: There is no mention of sanitizing or filtering the content of the files before the agent processes them.

security-review