implement
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a process for software development that aligns with industry best practices. It does not include any scripts, external dependencies, or network-enabled commands.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests specifications from local markdown files, creating a surface for potential instruction injection.
- Ingestion points: Acceptance criteria are read from
docs/specs/<slug>.md. - Boundary markers: None explicitly defined in the instructions.
- Capability inventory: The skill can perform file writes (implementation), run tests, and manage git branches.
- Sanitization: The skill does not mention validation or sanitization of the spec file content.
Audit Metadata