NYC

hypothesis-generation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Scientific Inquiry Protocol (SAFE): All files consist of markdown instructions, reference guidelines, and output templates. No executable scripts, binary files, or obfuscated payloads are present.\n- Indirect Prompt Injection Surface (LOW): The skill is designed to fetch and process external scientific literature, which creates a surface for indirect prompt injection. This is a common characteristic of research tools. 1. Ingestion points: Literature search results from PubMed and general web search in SKILL.md (Step 2). 2. Boundary markers: Absent; no specific delimiters are defined to isolate untrusted external content. 3. Capability inventory: The skill utilizes WebFetch and WebSearch capabilities to retrieve content. 4. Sanitization: Absent; the workflow does not specify methods for validating or escaping external input.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:52 PM