convex-create-component
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided feature descriptions to generate code, which is a standard pattern for this type of agent but constitutes an indirect injection surface. \n
- Ingestion points: User input regarding feature goals and repository structure (SKILL.md). \n
- Boundary markers: Not present in the provided instructions. \n
- Capability inventory: Generates files and runs development commands like npx convex dev. \n
- Sanitization: Relies on the agent's internal safety logic and user review of the resulting code. \n- [SAFE]: The skill enforces boundary security by directing developers to handle sensitive logic such as authentication and environment variable access in the main application layer rather than within the isolated component logic.
Audit Metadata