The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill provides instructions for standard development tasks such as npx create-next-app, npm install, and running local Python utilities (nextjs-init.py, turborepo-migrate.py). These commands are typical for project scaffolding and monorepo management.
[EXTERNAL_DOWNLOADS] (SAFE): The instructions reference downloading well-known, trusted packages from the npm registry (Next.js, Turborepo, RemixIcon). No unauthorized or suspicious remote sources were identified.
[CREDENTIALS_UNSAFE] (SAFE): Examples demonstrate the use of environment variables and CI/CD secrets (e.g., ${{ secrets.TURBO_TOKEN }}) following security best practices. No hardcoded API keys, tokens, or credentials are present.
[DATA_EXFILTRATION] (SAFE): Code snippets include network requests (e.g., fetch to api.example.com) that are clearly marked as placeholders for documentation purposes. No evidence of unauthorized data transmission was found.
[PROMPT_INJECTION] (SAFE): The documentation contains instructional text and code comments that do not attempt to override agent behavior or bypass safety filters.

web-frameworks