build-until-pass
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill identifies and executes local build and test commands (e.g., npm run build, cargo build) found within a project's configuration to verify code correctness.
- [PROMPT_INJECTION]: The skill analysis surface includes project configuration files and command error output. 1. Ingestion points: Phase 0 reads project files and Phase 1 captures shell error output. 2. Boundary markers: None. 3. Capability inventory: Local command execution and file modification. 4. Sanitization: Relies on behavioral constraints such as minimal diff requirements and a hard attempt cap to mitigate risks of indirect injection from malicious error messages.
Audit Metadata