Skills
Navigate
skills/ujang19/agent-skill-bedul/cloakbrowser/Gen Agent Trust Hub

cloakbrowser

Fail

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The install.sh script utilizes sudo to create and set permissions for an executable wrapper at /usr/local/bin/cloakbrowser-python.- [REMOTE_CODE_EXECUTION]: During the installation process, the install.sh script fetches and executes the installation script for the uv package manager from astral.sh.- [EXTERNAL_DOWNLOADS]: The skill downloads the uv package manager and a patched Chromium binary (approximately 200MB) from remote sources during setup.- [PROMPT_INJECTION]: The skill's ability to fetch and evaluate content from arbitrary URLs (scripts/fetch.py, scripts/eval-js.py, scripts/pdf.py, scripts/screenshot.py) creates an indirect prompt injection surface.
  • Ingestion points: URL parameters accepted by multiple bundled scripts.
  • Boundary markers: The scripts do not implement delimiters or warnings for processed content.
  • Capability inventory: The skill can execute arbitrary JavaScript in the browser context, generate PDF files, and capture screenshots.
  • Sanitization: No validation or sanitization of remote content is performed before returning data to the agent context.
Recommendations
  • HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 11, 2026, 07:04 PM