Skills
Navigate
skills/bmcszk/agent-skills/interactive-browser-testing/Gen Agent Trust Hub

interactive-browser-testing

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web content through browser navigation and utilizes AI to analyze screenshots of that content. This creates a surface where malicious UI elements or text could attempt to influence the agent's behavior.
  • Ingestion points: Web content accessed via browser_navigate in SKILL.md.
  • Boundary markers: None specified in the analysis prompts to separate page content from instructions.
  • Capability inventory: The agent has access to browser_evaluate for JavaScript execution and browser_file_upload for local file access.
  • Sanitization: No sanitization of visual or textual data from the browser is performed before AI processing.
  • [DYNAMIC_EXECUTION]: The skill provides the browser_evaluate tool which allows the execution of arbitrary JavaScript within the browser's context. While essential for complex browser testing, it is a high-privilege capability.
  • [COMMAND_EXECUTION]: The skill uses shell commands for environment preparation and tool connectivity.
  • Evidence: Usage of mkdir for results directories and mcpc connect for tool configuration in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 04:54 PM