website-scraper
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data fetched from external websites.
- Ingestion points: External HTML and data retrieved via the
web_fetchtool and the Pythonrequestslibrary in SKILL.md. - Boundary markers: The instructions lack specific delimiters or guardrails to prevent the agent from interpreting instructions potentially embedded in the scraped content.
- Capability inventory: The agent can execute Python code, write files to the local file system, and perform network requests.
- Sanitization: Data cleaning steps are limited to formatting (stripping HTML tags and whitespace) and do not include security filtering for malicious prompts.
- [COMMAND_EXECUTION]: The skill generates and executes dynamic Python scripts to implement complex scraping logic, pagination, and data export.
- [EXTERNAL_DOWNLOADS]: The skill fetches data from user-provided external URLs and installs common Python packages (requests, beautifulsoup4, lxml) from the official PyPI registry.
Audit Metadata