workflow-refactor

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). This workflow includes a runtime research step (Firecrawl) that can scrape public web pages (CallMcpTool with sources: [{ "type": "web" }] and firecrawl_scrape in SKILL.md), so outsider-authored free text from fetched URLs may be ingested into the agent’s LLM context via the tool results.