researcher

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's logic and commands are entirely consistent with its stated purpose of research and information synthesis. No obfuscation, unauthorized network calls, or credential theft patterns were detected.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests untrusted data from the internet via WebSearch and WebFetch tools as specified in SKILL.md. Mandatory Evidence Chain: 1. Ingestion points: WebSearch and WebFetch tools used for data collection. 2. Boundary markers: Absent in instructions. 3. Capability inventory: WebSearch, WebFetch, and Task (for sub-agent management). 4. Sanitization: No explicit filtering of fetched content mentioned. This surface is considered a functional requirement for a researcher agent and does not escalate the severity.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 04:31 AM