mcp-patterns

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

• Third-party content exposure detected (high risk: 0.80). This skill includes patterns that fetch and return external content (e.g., references/tool-patterns.md shows calling fetch_data(arguments["url"]) and references/auth-patterns.md uses httpx to call external APIs), so the agent can ingest untrusted public web content as part of its workflow, enabling indirect prompt injection.