corrinehu-kimi-searchzhihu

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the agent-browser package from the public NPM registry.
  • [COMMAND_EXECUTION]: The skill executes multiple shell commands to control a headless or headed Chrome browser, manage authentication states, and take screenshots of the user's desktop environment or specific browser windows.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process content from Zhihu search results, which are untrusted external inputs.
  • Ingestion points: Data enters the context through npx agent-browser open commands targeting Zhihu search result pages.
  • Boundary markers: No specific boundary markers or instructions to ignore embedded prompts are implemented in the search flow.
  • Capability inventory: The skill possesses capabilities for browser interaction (fill, click), session management (clear, list), and file system writes (screenshots).
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the web before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:09 PM