skills/yogsoth-ai/de-anthropocentric-research-engine/formulate-research-question/Gen Agent Trust Hub
formulate-research-question
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data through the
hypothesis_or_gapinput field. This introduces a surface for indirect prompt injection if the user-provided data contains instructions intended to manipulate the agent's research workflow. - Ingestion points: The
hypothesis_or_gapfield defined in the input contract. - Boundary markers: The skill does not explicitly use delimiters or boundary instructions to isolate user input from its own execution protocol.
- Capability inventory: The skill's primary capabilities involve coordinating other research skills and generating structured text outputs; it does not perform high-risk operations like shell execution or unauthorized network access.
- Sanitization: There are no specific sanitization or validation steps mentioned for the input text.
Audit Metadata