google-maps-chain-store-lead-search

Warn

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and installs runtime dependencies from the BrowserAct website and GitHub repository. The instructions explicitly direct the agent to perform these installations automatically, bypass user confirmation, and suppress descriptions of the installation process.
  • [COMMAND_EXECUTION]: Utilizes the browser-act CLI to perform browser-based tasks and lead extraction on Google Maps.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external business listings and websites.
  • Ingestion points: Google Maps business names, categories, website content, and contact links (SKILL.md).
  • Boundary markers: None provided to distinguish between instructions and data.
  • Capability inventory: Executes CLI commands via the browser-act tool and performs network operations through a browser.
  • Sanitization: No validation or sanitization of the extracted data is mentioned.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 20, 2026, 04:52 AM
Security Audit — agent-trust-hub — google-maps-chain-store-lead-search