browser-album-downloader

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires and installs well-known third-party libraries and software. The scripts/check_environment.sh script installs the playwright and requests Python packages via pip and the Chromium browser using the Playwright CLI tool. These are sourced from official and reputable registries.
  • [COMMAND_EXECUTION]: The skill includes shell scripts and Python templates intended for execution. scripts/check_environment.sh executes commands to verify dependencies and perform installations. The Python templates (templates/album_downloader.py and templates/scroll_collector.py) are designed to be run by the user in their local terminal to automate browser interactions and file downloads.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests and processes content from external, untrusted web pages.
  • Ingestion points: The scripts extract data from meta tags and img attributes (such as src, width, and height) from user-specified URLs using Playwright's page.evaluate and locator functions.
  • Boundary markers: None. The ingested content is used directly by the script's logic to determine download targets.
  • Capability inventory: The script has the capability to perform network operations via requests.get and write files to the project's downloads/ directory.
  • Sanitization: The safe_filename function in the downloader template uses regular expressions to sanitize identifiers extracted from the web before using them as local filenames.
  • [SAFE]: The Python templates utilize verify=False in requests.get calls, which disables SSL certificate verification. The code comments indicate this is a workaround for environment-specific proxy issues. While this is a security anti-pattern that could facilitate man-in-the-middle attacks, it is used here for downloading publicly accessible image content and is documented for the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 06:39 PM