deep-dive

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection as it fetches and processes untrusted external content from the web via the WebFetch and WebSearch tools. This is inherent to the skill's primary function of web-based research.
  • Ingestion points: Data enters the agent's context through content fetched by sub-agents using the WebFetch tool in Phase 2.
  • Boundary markers: The instructions do not define explicit delimiters (like XML tags) or specific directives for sub-agents to ignore instructions contained within the fetched web content.
  • Capability inventory: The skill utilizes Bash, Read, Write, WebSearch, WebFetch, Agent, and AskUserQuestion tools.
  • Sanitization: No explicit sanitization or filtering of the fetched content is performed prior to analysis and synthesis.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 09:52 AM