research-advisor

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a structured conversational guide for researchers. Analysis of all instructional content and reference documents shows no evidence of malicious patterns, prompt injection, or unauthorized data access.
  • [PROMPT_INJECTION]: The skill facilitates the analysis of user-provided research ideas and PubMed data. While this creates a theoretical surface for indirect prompt injection, it is standard for conversational research tool s using external data. Ingestion points: User research descriptions and results from PubMed searches. Boundary markers: No explicit delimiters are used to isolate user data from the core logic. Capability inventory: The skill suggests using PubMed search tool s for verification. Sanitization: No content validation or sanitization of external data is specified in the prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 12:27 PM
Security Audit — agent-trust-hub — research-advisor