The Agent Skills Directory

[SAFE]: The skill is a documentation-only asset consisting of markdown guidelines for code review. It does not include any commands for network exfiltration, sensitive file access, or unauthorized system modifications.
[NO_CODE]: No executable code, scripts, or package manifests (such as package.json or requirements.txt) are present in the skill, which removes the possibility of malicious runtime execution or supply chain attacks.
[PROMPT_INJECTION]: The instructions are task-oriented and lack any patterns designed to override agent behavior, bypass safety filters, or extract system prompts. The instructions explicitly encourage the detection of security vulnerabilities like unsafe input and sensitive data exposure in the code being reviewed.

review-feature