integrate-web-search

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill provides tools that ingest data from the public web, creating a surface for indirect prompt injection attacks.
  • Ingestion points: The fetch_page(url) tool retrieves content from external websites and returns it to the agent context.
  • Boundary markers: No explicit delimiters or boundary instructions are shown in the registration examples to prevent the agent from obeying instructions embedded in fetched web content.
  • Capability inventory: The agent is equipped with tools to browse the web, which could be influenced by malicious content if not carefully managed.
  • Sanitization: Content is processed with BeautifulSoup to remove scripts and styles, which reduces technical risk but does not filter the semantic content for malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 12:38 PM
Security Audit — agent-trust-hub — integrate-web-search