to-spec
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted conversation history and codebase data to generate specifications that are published to external tracking systems.
- Ingestion points: The skill synthesizes data from the "current conversation context and codebase understanding" (SKILL.md).
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its instructions and potentially malicious commands embedded in the conversation history.
- Capability inventory: The skill performs write operations by publishing synthesized specs to the project issue tracker (SKILL.md).
- Sanitization: No sanitization or validation of the ingested context is performed. However, the skill does include a human-in-the-loop verification step ("Check with the user that these seams match their expectations") which serves as a partial mitigation.
Audit Metadata