Swift Ecosystem

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow includes mcp__plugin_claude-code-home-manager_context7__query-docs / Context7 “get-library-docs” which fetches up-to-date Swift documentation from external sources and injects the fetched page text into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill declares a runtime Context7 MCP tool (mcp__plugin_claude-code-home-manager_context7__query-docs) that fetches external library docs by IDs such as "/swiftlang/swift" during execution and injects those documents into the agent's context to drive prompts, so it is a runtime external dependency that can directly control agent output.