rebuttal-writing
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external content, specifically reviewer comments and manuscript drafts.
- Ingestion points: As outlined in
SKILL.md, the workflow requires reading manuscripts, decision letters, and reviews provided by the user. - Boundary markers: The skill contains instructions to keep "reviewer quotations, author responses, and proposed manuscript text visibly distinct," which acts as a structural boundary to reduce the risk of the agent confusing external text for instructions.
- Capability inventory: The skill is limited to drafting and editing text. It does not define or request access to tools, network operations, or file system modifications.
- Sanitization: The skill does not implement programmatic sanitization or filtering of the input text, relying instead on structural guidelines for the agent.
Audit Metadata