The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown files providing documentation and architectural guidelines for React development. It does not contain executable scripts (.js, .py, .sh) or active platform configurations that trigger automated actions.
[REMOTE_CODE_EXECUTION]: The documentation references commands such as npx shadcn@latest add and npm install. These are standard developer tools and are presented as static usage examples rather than automated execution routines. The external resources mentioned (Radix UI, shadcn/ui, 21st.dev, Vercel) are well-known and reputable services in the frontend ecosystem.
[PROMPT_INJECTION]: No patterns of instruction overriding, safety filter bypasses, or 'developer mode' jailbreaks were found. The use of 'CRITICAL' and 'HIGH' labels in AGENTS.md refers to the importance of specific design principles (like accessibility) for developers, not security verdicts or safety claims.
[DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or unauthorized network operations were detected. All network references are standard documentation links or public package registries.
[OBFUSCATION]: The content was analyzed for hidden Unicode characters, homoglyphs, and Base64 encoded payloads. No malicious obfuscation techniques were found.

components-build