spec-first
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources that could contain instructions intended to override agent behavior.
- Ingestion points: The skill processes 'User request', 'repository' content, and 'existing issue/plan' as inputs in SKILL.md.
- Boundary markers: No delimiters or specific safety instructions were found to separate these external inputs from the agent's system prompt.
- Capability inventory: The skill has authority to create and modify repository artifacts and implementation files through delegated engines.
- Sanitization: No explicit content sanitization or instruction filtering was observed in the skill workflow for data retrieved from repositories or issues.
Audit Metadata