convex-design

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run npx tsc --noEmit and npx convex dev --once to verify the generated backend code. These are standard operations for the Convex development environment and are used to ensure code quality before deployment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input in the form of plain-English product ideas to guide code generation and verification.
  • Ingestion points: User-provided product descriptions and backend requirements entering the agent context.
  • Boundary markers: Absent; there are no specific instructions to delimit or ignore embedded instructions within user-provided product asks.
  • Capability inventory: The skill is capable of writing TypeScript files to the convex/ directory and executing shell commands through npx (TSC and Convex CLI).
  • Sanitization: Absent; the skill does not explicitly describe sanitization or filtering of the user's product requirements before using them to generate logic or code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:42 PM
Security Audit — agent-trust-hub — convex-design