competitor-analysis

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill is developed by 'browserbase' and integrates exclusively with their official 'browse' CLI and Cloud APIs. All tool invocations and network requests target the vendor's own infrastructure.
  • [COMMAND_EXECUTION]: The research pipeline relies on several local Node.js scripts (e.g., compile_report.mjs, capture_screenshots.mjs) and the 'browse' CLI tool. These executions are restricted to the skill's operational directory and intended research output paths.
  • [EXTERNAL_DOWNLOADS]: The skill requires the 'browse' CLI tool, which is an official package from the vendor's well-known service. This dependency is standard for the skill's functionality and originates from a trusted source.
  • [SAFE]: Indirect Prompt Injection Surface. The skill fetches and processes content from external websites found during the discovery phase. This is a standard functional requirement for a web-research agent. The skill implements basic sanitization via the 'stripHtml' utility and uses structured data parsing to mitigate common injection risks associated with untrusted web content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 08:50 PM
Security Audit — agent-trust-hub — competitor-analysis