marketing-community
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The shared-patterns.md file includes instructions to install the agent-browser tool from the trusted vercel-labs repository on GitHub.
- [COMMAND_EXECUTION]: The skill contains commands for setting up environment dependencies using npm and npx, targeting a trusted organization's repository for browser automation.
- [PROMPT_INJECTION]: The skill facilitates auditing live website URLs, which establishes a surface for indirect prompt injection. This is a standard risk for agents performing web-based research and is managed by system-level guardrails.
Audit Metadata