What problem does it solve? The Vercel AI SDK evolves rapidly, and outdated training data leads to deprecated APIs, wrong parameter names, and broken code. This Skill forces verification against bundled docs and source code so generated AI features actually compile and run. ## Core Features & Use Cases - Verified API Guidance: Searches node_modules/ai/docs and source code, falling back to ai-sdk.dev, instead of relying on memory. - Agent & Chatbot Building: Creates ToolLoopAgent-based agents with type-safe useChat consumption via InferAgentUIMessage. - Migration & Error Fixing: Resolves common breaking changes like maxTokens to maxOutputTokens, parameters to inputSchema, and useChat state management updates. - Use Case: A developer asks to add a streaming chatbot to a Next.js app; the Skill checks the installed ai package docs, fetches current model IDs from the Vercel AI Gateway, and generates type-safe server and client code that passes typecheck. ## Quick Start Ask the assistant to build a streaming chat endpoint using the AI SDK with the latest Anthropic model and have it verify the code against the installed package docs.