water-research

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides heuristic advice and scope boundaries for academic publishing. It does not include any scripts, commands, or network exfiltration patterns.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied research descriptions which constitutes an indirect prompt injection surface. 1. Ingestion points: Manuscript text and research descriptions analyzed against criteria in SKILL.md. 2. Boundary markers: None present. 3. Capability inventory: The skill has no dangerous capabilities such as shell access, file writing, or network tools. 4. Sanitization: None present. The risk is considered safe as the skill lacks tools to perform actions outside the dialogue context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 11:33 AM
Security Audit — agent-trust-hub — water-research