The Agent Skills Directory

[SAFE]: The skill's instructions are focused entirely on static code analysis and reporting. It does not request or use any tools that would allow for network access, file system modifications, or system changes.- [PROMPT_INJECTION]: The instructions do not contain any patterns intended to bypass safety filters, override system prompts, or extract internal instructions. The language is purely operational and instructional.- [REMOTE_CODE_EXECUTION]: No remote code execution or external script fetching was detected. The skill does not define any external dependencies or package installation steps.- [DATA_EXFILTRATION]: While the skill reviews codebases, there is no logic or instruction to send the analyzed data to external servers or non-whitelisted domains.- [COMMAND_EXECUTION]: The skill does not utilize shell commands or subprocess execution within its logic.

review-codebase