skills/fengjiehzi/skills/browser-use/Gen Agent Trust Hub

browser-use

Warn

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill provides commands that access and potentially exfiltrate sensitive information. The cookies get command exports browser cookies, and the upload command can be used to send local files to remote websites.- [COMMAND_EXECUTION]: The skill relies on the Bash tool to execute browser-use CLI commands for all browser operations.- [EXTERNAL_DOWNLOADS]: The uvx browser-use install command triggers the download and installation of the Chromium browser binary from external remote sources.- [REMOTE_CODE_EXECUTION]: The skill uses uvx to fetch and execute the browser-use package from a remote package registry at runtime.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted content from the web. Ingestion points include uvx browser-use state and uvx browser-use extract. There are no explicit boundary markers or instructions to isolate web content from the agent's command processing. The agent has significant capabilities including shell execution, file reading, and cookie management, which could be abused if malicious web content influences its behavior. No evidence of sanitization was found.- [DYNAMIC_EXECUTION]: The uvx browser-use eval command allows for the execution of arbitrary JavaScript code within the context of a loaded web page, facilitating runtime code injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 19, 2026, 06:45 AM