browser-automation

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the agent-browser CLI tool to perform browser automation tasks, including navigation, element interaction, and JavaScript execution within the page context.
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection. By allowing the agent to fetch and process content from external websites, there is a risk that malicious instructions embedded in those pages could influence the agent's subsequent actions.
  • Ingestion points: Web content retrieved via agent-browser navigate, snapshot, and execute commands (SKILL.md).
  • Boundary markers: None provided in the instructions to help the agent distinguish between untrusted web data and authoritative instructions.
  • Capability inventory: The agent is granted access to sensitive tools such as Bash, Read, and Write (SKILL.md frontmatter).
  • Sanitization: The skill does not define any validation or sanitization logic for the data extracted from external URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 12:01 PM