scrapling

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the scrapling package from the Python Package Index (PyPI) using uv tool install. This is a public tool for web scraping and browser automation with official documentation and source code available on GitHub (github.com/D4Vinci/Scrapling).
  • [COMMAND_EXECUTION]: The skill invokes the scrapling CLI to perform network operations and save scraped content to the local file system. It includes options for customizing headers, cookies, and browser fingerprints, as well as an --executable-path option to specify a custom Chromium binary for browser automation.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external websites, which serves as an attack surface for indirect prompt injection. The skill documentation follows best practices by mandating the use of the --ai-targeted flag.
  • Ingestion points: Content is fetched from arbitrary URLs provided to the extract command (e.g., SKILL.md).
  • Boundary markers: The skill writes output to temporary files (e.g., /tmp/page.md) before the agent reads them.
  • Capability inventory: The skill can perform HTTP requests, bypass anti-bot protections, automate browsers, and write files to disk.
  • Sanitization: The --ai-targeted flag is specifically included to sanitize content for AI consumption by removing noisy/hidden content, stripping comments and zero-width characters, and blocking ads.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 01:01 PM