Chrome Bridge Automation
Pass
Audited by Gen Agent Trust Hub on Mar 20, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill utilizes
npx @midscene/web@1to obtain and run the automation CLI from the official NPM registry. - [COMMAND_EXECUTION]: It uses the
Bashtool to interact with the browser automation environment. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interprets instructions based on screenshots of live websites. Malicious content on a page could attempt to subvert the agent's logic or trigger unauthorized actions.
- Ingestion points: Captures screenshots of websites visited during automation (SKILL.md).
- Boundary markers: No clear markers exist to separate website content from the agent's core instructions.
- Capability inventory: Bash shell access, file read/write for screenshots, and network access via the browser.
- Sanitization: No sanitization or filtering of visual data is performed.
- [EXTERNAL_DOWNLOADS]: The skill connects to external AI model providers (such as Google, Aliyun, or Volcengine) to process vision-based tasks.
Audit Metadata