spec-kit-implement
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill does not exhibit any malicious behaviors such as data exfiltration, unauthorized network access, or obfuscation.
- [PROMPT_INJECTION]: The skill is designed to ingest instructions from external files, presenting a surface for indirect instructions.
- Ingestion points: Reads implementation tasks from specs/changes//tasks.md and design plans from plan.md.
- Boundary markers: None identified; the skill directly follows file content to drive the development loop.
- Capability inventory: The skill can modify local project files (tasks.md, spec.md) and generate implementation code based on file contents.
- Sanitization: None; the skill assumes the integrity of the project's design and task files.
Audit Metadata