agent-browser
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to interact with and process data from external web interfaces, including accessibility snapshots, console logs, and network requests. This creates an attack surface where malicious content on a target website could attempt to influence the agent's behavior.
- Ingestion points: Web content retrieved via
agent-browser open, accessibility snapshots, and console/network logs. - Boundary markers: None explicitly mentioned in the skill instructions to delimit external content.
- Capability inventory: Execution of
agent-browserCLI commands for navigation, interaction, and evidence capture. - Sanitization: The skill does not specify methods for sanitizing or escaping the data retrieved from the browser before processing.
- [COMMAND_EXECUTION]: The skill's core functionality relies on executing the
agent-browserCLI tool (e.g.,agent-browser open,agent-browser skills get). These are standard operations for the skill's intended purpose of UI testing.
Audit Metadata