spec-first
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- COMMAND_EXECUTION (LOW): The skill instructions explicitly direct the agent to generate and execute shell commands for verification (e.g.,
npm run build && npm test,open docs/index.html). - Ingestion points: The agent creates and then reads from
.agents/TODOS/[feature-name].mdfiles. - Boundary markers: None present in the provided templates to prevent instructions in the spec or todo files from being interpreted as agent commands.
- Capability inventory: The workflow assumes the agent can execute shell commands, run tests, and open files.
- Sanitization: No sanitization is mentioned; the agent is encouraged to 'run verification command, paste actual output back'.
Audit Metadata