The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown content and checklists meant to guide an agent during code reviews. No malicious command execution, network exfiltration, or persistence patterns were found.
[SAFE]: Code snippets included in the documentation are clearly marked as examples of good and bad practices (e.g., TypeScript and AI prompt patterns) and are not intended for execution by the agent.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted code snippets provided by users. It uses restrictive, read-only tools (Read, Glob, Grep) and lacks any capabilities for file writing, network transmission, or dynamic execution. The checklist explicitly includes a security check for prompt injection in the code being reviewed, which is a defensive best practice.

code-review-checklist