simulate
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute standard development commands (e.g.,
bun run build,bun test,bunx better-convex). These commands are used within the context of building and verifying a generated demo application in an isolated environment. - [SAFE]: The skill includes explicit safety instructions that prevent the agent from reading or modifying the existing repository, mandating all operations occur within a temporary directory (
/tmp/simulation-1). - [SAFE]: The use of the
better-convextool is consistent with the skill's purpose of testing that specific framework onboarding flow, aligning with vendor-expected behavior.
Audit Metadata