convex-conventions

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill establishes developer conventions and best practices for Convex application development without introducing security risks or malicious behavior.
  • [COMMAND_EXECUTION]: The instructions reference standard development CLI tools such as bun pm view, npx convex, and bun run typecheck. These commands are standard within the development workflow and operate on local project files.
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch information from the official docs.convex.dev domain and recommends official Convex components from the @convex-dev namespace. These resources originate from the primary platform vendor and are well-known within the ecosystem.
  • [PROMPT_INJECTION]: The skill includes a 'Live docs are authoritative' rule that instructs the agent to override the skill's own content if it conflicts with official documentation. This is a functional constraint to ensure code correctness and safety rather than a malicious attempt to bypass guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 08:49 AM
Security Audit — agent-trust-hub — convex-conventions