python-sdk-best-practices

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: All external domains (brightdata.com) and software packages (brightdata-sdk) referenced in the documentation are official resources belonging to the authenticated author, 'brightdata'.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external websites and search engines through its scraping and SERP APIs. This represents a potential surface for indirect prompt injection where malicious content on scraped pages could attempt to influence agent behavior.
  • Ingestion points: Data entering through scrape_url, client.scrape.*, client.search.*, and client.datasets.download methods.
  • Boundary markers: The documentation does not specify boundary markers for the data returned to the agent context.
  • Capability inventory: The skill provides capabilities for data export to local files (JSON/CSV), browser automation via Playwright, and general network connectivity.
  • Sanitization: There is no guidance on sanitizing or escaping the scraped content before the agent processes it.
  • [CREDENTIALS_UNSAFE]: The documentation provides instructions for managing API tokens and browser credentials, highlighting best practices such as using environment variables (BRIGHTDATA_API_TOKEN) and .env files to avoid hardcoding secrets in code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:45 PM
Security Audit — agent-trust-hub — python-sdk-best-practices