drupal-tutorial-video

Pass

Audited by Gen Agent Trust Hub on Oct 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the Montserrat font from the official Google Fonts repository on GitHub during the preflight setup phase.
  • [EXTERNAL_DOWNLOADS]: Installs the agent-browser and awaz packages from the npm registry to facilitate browser automation and narration generation.
  • [COMMAND_EXECUTION]: Extensively utilizes ddev exec to run shell commands on the host and within the web container, managing processes like ffmpeg for screen capture and xdotool for cursor movement.
  • [DYNAMIC_EXECUTION]: Dynamically generates and evaluates JavaScript expressions within the browser context to identify UI elements and retrieve on-screen coordinates.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Reads module info files, READMEs, and browser content from the ddev project to learn setup steps and generate storyboards (e.g., in SKILL.md and ui.sh).
  • Boundary markers: The skill does not employ explicit delimiters or instructions to ignore embedded instructions when reading external documentation.
  • Capability inventory: The skill possesses broad capabilities, including system package installation, file system access, and the ability to execute arbitrary commands within the container.
  • Sanitization: Employs base64 encoding when passing text to typing functions to prevent shell injection, but does not perform comprehensive sanitization of ingested documentation content.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 18, 2026, 04:37 PM