to-spec
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (conversation history and codebase) to generate and publish specifications.
- Ingestion points: The skill explicitly uses the current conversation context and performs codebase exploration (SKILL.md).
- Boundary markers: There are no defined delimiters or instructions provided to the agent to differentiate between genuine data and malicious instructions that might be embedded in the codebase or chat history.
- Capability inventory: The agent is empowered to read the repository and write/publish content to a project issue tracker (SKILL.md).
- Sanitization: The instructions do not specify any validation or sanitization steps for the data gathered from the repo or conversation before it is published externally.
Audit Metadata