firecrawl

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of ingesting untrusted web content.
  • Ingestion points: scripts/firecrawl_api.py and scripts/deepwiki.sh retrieve content from external URLs provided at runtime, which is then presented to the AI agent.
  • Boundary markers: The skill does not implement explicit delimiters (e.g., XML tags or unique markers) or specific system instructions to warn the agent to ignore instructions found within the scraped content.
  • Capability inventory: The skill possesses significant capabilities, including performing complex web searches, scraping any URL, and writing data to the local filesystem (via deepwiki.sh or shell redirection).
  • Sanitization: While scripts/filter_web_results.py provides useful markdown-level filtering (sections, keywords, and truncation), it does not sanitize the text for embedded instructions or malicious prompt segments.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 11:30 PM