browser-test
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes data from external websites, creating an attack surface for indirect prompt injection if an agent navigates to a malicious URL.
- Ingestion points: External content enters the agent context through the browser automation tools described in
SKILL.md(e.g.,mcp__claude-flow__browser_open). - Boundary markers: There are no instructions or delimiters in the prompt to separate retrieved web data from the agent's internal instructions.
- Capability inventory: The skill's environment includes tools for complete browser control, JavaScript execution (
browser_eval), and a shell interface. - Sanitization: The skill does not implement validation or escaping for the content retrieved from external pages.
- [COMMAND_EXECUTION]: The skill's workflow includes the use of
mcp__claude-flow__browser_evalfor executing custom JavaScript assertions within the browser context, which is a form of dynamic code execution intended for UI validation.
Audit Metadata