The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns, obfuscation, or unauthorized data access were detected. The skill provides boilerplate code and educational content for a legitimate technology stack.
[EXTERNAL_DOWNLOADS] (SAFE): The skill references standard initialization commands (e.g., npm create convex@latest) and well-known packages in the React/Convex ecosystem. These are standard development practices for the described stack.
[INDIRECT_PROMPT_INJECTION] (SAFE): While the skill provides templates for building AI agents in references/09-agents.md that ingest user input, these are purely educational examples for the user's own application and do not represent a vulnerability within the skill itself. The examples include standard sanitization patterns like JSON.stringify for tool outputs.

convex-tanstack