research
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to process data from external primary sources, creating a surface for potential indirect prompt injection. 1. Ingestion points: The agent reads official documentation, source code, and first-party APIs. 2. Boundary markers: The instructions lack explicit delimiters or directions to treat external content as data only. 3. Capability inventory: The task uses network access for research and filesystem write access to save findings. 4. Sanitization: No content validation or sanitization of the research data is defined before it is written to the repository.
Audit Metadata