chrome-replay

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to parse and interpret instructions from external Chrome DevTools Recorder files (JSON and JS), creating a surface for indirect prompt injection.
  • Ingestion points: Processes recording files provided by users or external sources as described in SKILL.md.
  • Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the processed recording data.
  • Capability inventory: The skill utilizes the Chrome DevTools MCP to perform actions such as navigate_page, click, type_text, press_key, and evaluate_script based on the recording content.
  • Sanitization: The instructions do not specify any sanitization or validation of the recording steps before execution.
  • [EXTERNAL_DOWNLOADS]: The context.md file contains instructions for the user to download and run the chrome-devtools-mcp package using npx during setup.
  • [COMMAND_EXECUTION]: The skill documentation provides technical guidance involving command-line utilities:
  • context.md includes a troubleshooting command to terminate browser processes using pkill.
  • SKILL.md suggests using jq and python to inspect and process recording files locally.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 09:20 AM