firecrawl-sdk-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill correctly demonstrates using environment variables for the Firecrawl API key, avoiding hardcoded secrets in source code.
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for building web scrapers that ingest data from external, untrusted sources, creating a potential surface for indirect prompt injection. 1. Ingestion points: The
scrapePage(TS) andscrape_with_retry(Python) functions ingest content from external URLs. 2. Boundary markers: The code patterns do not include explicit prompt boundary markers or 'ignore' instructions for scraped text. 3. Capability inventory: The patterns involve the Firecrawl SDK which performs network requests. 4. Sanitization: The skill recommends using Zod for response validation, ensuring structural integrity of the ingested data.
Audit Metadata