firecrawl-demo-walkthrough
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to browse and scrape content from third-party websites using the Firecrawl browser. This introduces a surface for indirect prompt injection, as malicious instructions embedded on a target website could potentially influence the agent's behavior during the walkthrough.
- Ingestion points: Untrusted data enters the agent context via the Firecrawl browser while snapshotting and scraping product pages and documentation (SKILL.md).
- Boundary markers: No specific delimiters or "ignore instructions" warnings are defined to separate the scraped website content from the agent's operational instructions.
- Capability inventory: The skill is capable of navigating flows, snapshotting pages, and documenting findings based on external content (SKILL.md).
- Sanitization: There is no evidence of sanitization, escaping, or validation of the content retrieved from the target URLs.
Audit Metadata