firecrawl-hardened
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from the web, which creates a surface for indirect prompt injection. This is addressed by defensive guardrails in SKILL.md that instruct the agent to maintain strict boundaries.
- Ingestion points: Found in scripts/search.py, scripts/scrape.py, and scripts/crawl.py.
- Boundary markers: Explicit instructions are provided to treat all scraped data as untrusted.
- Capability inventory: The scripts are limited to data retrieval and do not possess file system write or code execution capabilities.
- Sanitization: Instructions require the agent to redact credential-like patterns.
- [EXTERNAL_DOWNLOADS]: The skill performs network operations to the Firecrawl API at api.firecrawl.dev. This interaction is essential for the skill's operation and targets a well-known service domain.
Audit Metadata