create-research-questions
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources such as ticket files and collateral materials to generate research questions and guide further agent actions.
- Ingestion points: The agent is instructed to read all @-mentioned files, including ticket and collateral files, which can contain instructions from external actors.
- Boundary markers: There are no specific delimiters or instructions to treat the file content strictly as data, increasing the risk that the agent follows embedded instructions.
- Capability inventory: The skill possesses the ability to read and write files within the vendor-specific .humanlayer directory, execute directory listings, and invoke specialized research agents.
- Sanitization: No explicit sanitization or filtering of the ingested content is performed before it is used to generate the research questions.
Audit Metadata