The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes npx react-doctor@latest for automated codebase auditing and npx expo install or npx @next/codemod for project maintenance. These commands are standard developer operations within the React ecosystem and are used here for their intended purpose of improving code quality and performance.
[EXTERNAL_DOWNLOADS]: The skill fetches rule recipes from https://www.react.doctor/prompts/rules/ and references trusted repositories like millionco/react-doctor and Quegenx/agent-skills. These downloads provide necessary context for audit plans and originate from well-known or vendor-owned sources.
[INDIRECT_PROMPT_INJECTION]: While the skill involves auditing untrusted codebases, it implements a critical safeguard (Hard Rule 4) explicitly instructing the agent to treat repository content as data rather than instructions, mitigating potential injection vectors.
[SAFE]: The instructions consistently advocate for secure coding practices, such as authenticating Server Actions, avoiding secret exposure in client bundles, and using React's experimental taint APIs for data security.

react