The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a shell script scripts/sync-to-agents.sh designed to install or synchronize the skill files to local agent directories (e.g., ~/.claude/skills/). The script uses standard utilities like rsync, mkdir, and ln to manage the skill's deployment.
[EXTERNAL_DOWNLOADS]: The documentation (README.md) suggests installing the skill via npx skills, referencing a public utility repository from Vercel Labs. This is a reference to a well-known development service.
[PROMPT_INJECTION]: As a code auditing tool, the skill is designed to process untrusted data such as source code and pull request comments. This constitutes an indirect prompt injection surface. The skill mitigates this by instructing the agent to use a rigorous, multi-role (Security, Performance, UX, etc.) audit framework with cross-agent verification steps to maintain accuracy and safety.

audit-code