nature-ecology-and-evolution
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted data (manuscripts and conceptual claims) to evaluate their fit for a specific journal. It possesses capabilities to read local project files (e.g.,
../../resources/source-basis.md) and perform web searches for live instructions. - Ingestion points: User-provided manuscript text, target venue names, and conceptual claims in
SKILL.mdsections like 'When to trigger'. - Boundary markers: None explicitly defined in the instructions to separate manuscript content from agent instructions.
- Capability inventory: File read access to relative paths and web search capabilities for retrieving live guidelines.
- Sanitization: The skill lacks explicit sanitization or validation logic for the content of the manuscripts it processes.
Audit Metadata