Output AI工作流框架

Build AI using AI

Output is the first framework designed for AI coding agents. The entire codebase is structured so Claude Code can scaffold, plan, generate, test, and iterate on your workflows. Every workflow is a folder — code, prompts, tests, evals, traces, all together. Your agent reads one folder and has full context.

Quick Start

Requirements:

• Node.js 20+
• Docker Desktop
• An LLM API key (e.g. Anthropic)

Scaffold a project and add your API key to .env (ANTHROPIC_API_KEY=sk-ant-...):

npx @outputai/cli init
cd <project-name>

Start the full development environment — Temporal server, API server, a worker with hot reload, and the Temporal UI at http://localhost:8080:

npx output dev

Run your first workflow and inspect the execution:

npx output workflow run blog_evaluator paulgraham_hwh

npx output workflow debug <workflow-id>

For the full getting started guide, see the documentation.

Example Workflows

Production-ready workflows you can run locally, learn from, and fork — all from the output-examples gallery:

Browse the full gallery at output.ai/gallery.

Configuration

For production configuration and advanced settings (LLM providers, Temporal Cloud, tracing, and more), see the operations docs.

SDK Packages

Workflows

Orchestration layer — deterministic coordination logic, no I/O.

// src/workflows/research/workflow.ts
workflow({
  name: 'research',
  fn: async (input) => {
    const data = await gatherSources(input);
    const analysis = await analyzeContent(data);
    const quality = await checkQuality(analysis);
    return quality.passed ? analysis : await reviseContent(analysis, quality);
  }
});