principle-build-the-lever
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines engineering principles to improve code quality and reviewability. It contains no executable code, network calls, or sensitive data access. It promotes best practices like manual verification and deterministic automation.
- [DYNAMIC_EXECUTION]: The skill advises agents to generate and run scripts for tasks. This is consistent with its role as a developer assistant tool and does not involve executing external untrusted payloads or remote code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes task instructions to determine if a 'lever' (tool) should be built.
- Ingestion points: User-provided task descriptions and context (SKILL.md).
- Boundary markers: None explicitly defined in this skill file.
- Capability inventory: Capability to write and execute scripts for automation as described in the instructions.
- Sanitization: Not explicitly addressed, but the principle emphasizes 'rerunnable checks' and manual first-unit recipes to ensure correctness.
Audit Metadata