brainstorm
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by instructing the agent to process data from the working directory.
- Ingestion points: The agent is directed to 'check out the current state of the project in our working directory' (SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in the project files are provided.
- Capability inventory: The skill utilizes file system read capabilities to analyze existing project code and documentation.
- Sanitization: No sanitization or validation of the directory content is specified.
Audit Metadata