paper-writing
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes 'research findings' from potentially untrusted sources to generate technical drafts. While standard agent guardrails apply, there is a risk that malicious instructions embedded in research data could influence the output draft.
- Ingestion points: Research findings provided by the user or external tools referenced in the skill description (SKILL.md).
- Boundary markers: None explicitly defined in the provided skill instructions to delimit untrusted data.
- Capability inventory: The skill specifies writing output to the
papers/directory. - Sanitization: No specific sanitization, validation, or filtering logic is defined for the external research content.
Audit Metadata