Skip to main content
Middleware integration for the Vercel AI SDK. Adds cascade routing to AI SDK applications with streaming support.

​
Install

npm install @cascadeflow/vercel-ai

​
createChatHandler

Creates a request handler for AI SDK chat endpoints. 
import { createChatHandler } from '@cascadeflow/vercel-ai';
import { CascadeAgent } from '@cascadeflow/core';

const agent = new CascadeAgent({
  models: [
    { name: 'gpt-4o-mini', provider: 'openai', cost: 0.000375 },
    { name: 'gpt-4o', provider: 'openai', cost: 0.00625 },
  ],
});

const handler = createChatHandler(agent, {
  protocol: 'data',
  tools,
  toolHandlers,
  maxSteps: 5,
});

​
Options

interface ChatHandlerOptions {
  protocol: 'data' | 'ui';             // AI SDK stream protocol
  tools?: ToolDefinition[];             // Tool definitions
  toolHandlers?: Record<string, Function>; // Server-side tool execution
  toolExecutor?: Function;              // Universal tool executor
  maxSteps?: number;                    // Multi-step tool loop limit
  forceDirect?: boolean;                // Skip cascade, use verifier
  allowOverrides?: string[];            // Request-level override keys
  overrideSecret?: string;              // Shared secret for overrides
}

​
Features

  • AI SDK v4 data stream and v5/v6 UI streams
  • useChat multi-turn support
  • parts message format (AI SDK v6)
  • Tool call streaming visibility
  • Server-side tool execution loops
  • Multi-step controls
  • Cascade decision stream parts
  • Request-level overrides with allowlist