chrome-devtools

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references npx chrome-devtools-mcp@latest, which downloads and executes the MCP server from the public npm registry. This is a standard method for utilizing this tool.
  • [COMMAND_EXECUTION]: The documentation mentions configuring the browser and MCP server via command-line interface (CLI) arguments during setup.
  • [DYNAMIC_EXECUTION]: The evaluate_script tool enables the execution of JavaScript within the browser environment to retrieve data or interact with elements not covered by standard automation tools.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of content from external web pages, which represents an untrusted data source.
  • Ingestion points: Web page content is ingested through take_snapshot, take_screenshot, and evaluate_script.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are specified in the provided markdown.
  • Capability inventory: The skill can perform actions such as click, fill, navigate_page, and evaluate_script (JavaScript execution).
  • Sanitization: There is no mention of sanitization or filtering for the data retrieved from web pages before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 05:49 AM
Security Audit — agent-trust-hub — chrome-devtools