context7

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

• Third-party content exposure detected (medium risk: 0.30). At runtime, the workflow uses user-provided query in /query-docs (after resolve-library-id) and then prints the tool’s returned item.text from Context7 MCP server output, meaning outsider-authored free text (e.g., from external docs/examples) is ingested without selecting a specific trusted item.