Package Exports
- @ai-sdk/devtools
Readme
AI SDK DevTools
A local development tool for debugging and inspecting AI SDK applications. View LLM requests, responses, tool calls, and multi-step interactions in a web-based UI.
Note: This package is experimental and intended for local development only. Do not use in production environments.
Installation
npm install @ai-sdk/devtools
# or
pnpm add @ai-sdk/devtoolsRequirements
- AI SDK v7 canary (
ai@canary) - Node.js compatible runtime
Usage
1. Register the telemetry integration
Register DevToolsTelemetry globally so it captures all AI SDK calls:
import { registerTelemetry } from 'ai';
import { DevToolsTelemetry } from '@ai-sdk/devtools';
registerTelemetry(DevToolsTelemetry());Telemetry is enabled automatically once an integration is registered:
import { generateText } from 'ai';
const result = await generateText({
model: yourModel,
prompt: 'What cities are in the United States?',
});You can also pass the integration to individual calls instead of registering it globally:
import { streamText } from 'ai';
import { DevToolsTelemetry } from '@ai-sdk/devtools';
const result = streamText({
model: yourModel,
prompt: 'Hello!',
telemetry: {
integrations: [DevToolsTelemetry()],
},
});2. Run the viewer
npx @ai-sdk/devtoolsOpen http://localhost:4983 to view your AI SDK interactions.
If you are using a monorepo, start DevTools from the same workspace where your AI SDK code runs.
How it works
The DevToolsTelemetry integration hooks into the AI SDK telemetry lifecycle to
capture generateText, streamText, generateObject, and streamObject calls.
It captures:
- Input parameters and prompts
- Output content and tool calls
- Token usage and timing
- Raw provider data
Data is stored locally in a JSON file (.devtools/generations.json) and served through a web UI.
Data flow
AI SDK call -> DevToolsTelemetry -> JSON file -> Hono API -> React UIKey concepts
- Run: A complete multi-step AI interaction, grouped by initial prompt
- Step: A single LLM call within a run
Development
pnpm install
pnpm dev # Start dev server at http://localhost:5173License
MIT