The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions found that attempt to bypass safety filters or override the agent's default behavior.
[Data Exposure & Exfiltration] (SAFE): The skill correctly demonstrates using process.env for API keys and database credentials. It identifies that secrets should never be exposed to the client and uses non-sensitive placeholders like sk-xxx in examples.
[External Downloads] (SAFE): The skill references legitimate, well-known developer tools and libraries including eas-cli and @libsql/client/web for standard integration workflows.
[Indirect Prompt Injection] (SAFE): An example for an AI proxy endpoint in app/api/ai+api.ts demonstrates passing user input to an LLM. While this is a common vulnerability surface, it is presented here as an educational code pattern for developers.
[Command Execution] (SAFE): Documentation includes curl examples and npx expo commands intended for local development and testing by the user.

api-routes