paper-revision
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, potentially untrusted content such as academic drafts and reviewer comments.\n
- Ingestion points: The instructions guide the agent to process raw text pasted by the user for revision or response drafting (SKILL.md).\n
- Boundary markers: The skill lacks explicit delimiters or instructions to treat the user-provided text strictly as data to prevent potential subversion of the agent's instructions.\n
- Capability inventory: The skill's operations are limited to natural language analysis and text generation. It does not utilize sensitive tools, file system access, or network capabilities.\n
- Sanitization: No mechanisms for sanitizing or validating the input text are described in the instructions.
Audit Metadata