demineur4356435

demineur4356435 / ts-ai-agent-orchestrator

Public

Multi-agent orchestration in TypeScript: DAG workflow scheduler, Zod schemas, Vercel AI SDK generateText/streamText, hooks & streaming. For research pipelines & agentic automation.

agentic-ai ai-agents anthropic dag llm
20
0
100% credibility
Found Apr 19, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A TypeScript library for building and running workflows of multiple AI agents arranged in dependency graphs, supporting parallel execution, structured inputs/outputs, and streaming from providers like OpenAI, Anthropic, and Ollama.

How It Works

1
🔍 Discover AI Team Tool

You stumble upon this handy kit that lets groups of smart AI helpers team up to tackle big tasks like research or analysis.

2
📥 Set Up Your Kit

Download the files to your computer and prepare everything so it's ready to go.

3
🔗 Link AI Brains

Connect to a smart AI service, like one from OpenAI or a local one, so your helpers can think and chat.

4
📝 Pick a Sample Task

Choose a ready-made example, like a researcher that gathers facts and a summarizer that boils them down.

5
💬 Ask Your Question

Type in something you're curious about, such as the latest trends in a topic, and launch the team.

6
⚡ Watch Them Work

See your AI team jump into action, with some helpers working side by side and passing info along smoothly.

🎉 Enjoy Smart Results

Get back a neat, organized answer or summary that feels like having a team of experts at your fingertips.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ts-ai-agent-orchestrator?

This TypeScript library orchestrates multi-agent AI workflows as directed acyclic graphs (DAGs), letting you define agents with Zod input/output schemas, wire dependencies, and run parallel layers via the Vercel AI SDK's generateText and streamText. It handles structured outputs, token streaming, and hooks for events like agent start/complete, supporting OpenAI, Anthropic, and local Ollama models. Developers get a type-safe way to chain LLM agents for tasks like research pipelines or agentic automation, with a CLI for quick runs and a REPL for interactive testing.

Why is it gaining traction?

In a sea of Python tools like LangGraph multi-agent GitHub repos or n8n-style multi-agent orchestration, this stands out as a lightweight, native TypeScript option—no transpilers or heavy runtimes needed. The DAG scheduler rejects cycles upfront, parallel execution speeds up workflows, and built-in streaming with token events makes real-time UIs straightforward. It's an open-source multi-agent orchestration platform that pairs well with GitHub Copilot for quick agent prototyping, minus the vendor lock-in of Microsoft Copilot Studio.

Who should use this?

AI engineers building research pipelines or multi-step Q&A chains in TypeScript apps. Backend devs automating document analysis or code-review workflows with LLMs. Teams experimenting with agentic systems who want local Ollama support and Zod-typed inputs without switching to Python multi-agent GitHub Copilot alternatives.

Verdict

Grab it for prototyping multi-agent orchestration—solid docs, CLI/REPL, and tests make it dead simple to start, despite 20 stars and 1.0% credibility signaling early maturity. Scale up once your workflow stabilizes; it's not battle-tested for production yet.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.