Build AI Agents Visually.

Open-source platform for LangChain & LangGraph workflows. Drop agents on a canvas, connect them, deploy.

Making agentic AI accessible.

Get Started on GitHub Watch Demo

From idea to working agent is too many steps.

Graph definitions, state management, prompt engineering, tool integration. By the time you've wired it all together, you've lost the idea.

LangConfig gives you a visual canvas. Drop agents, connect workflows, watch them think in real-time. Export production-ready code when you're done.

Watch Agents Think

Tool selection, reasoning, and outputs stream live. Debug by seeing, not reading logs.

  • ✓ Live execution streaming
  • ✓ Visual error detection
  • ✓ Interactive tool inspection

Canvas to Production

Export as a Python package with Streamlit UI, or share as JSON anyone can import.

  • ✓ One-click Python export
  • ✓ Streamlit UI included
  • ✓ JSON configuration sharing
LangConfig Workflow Export
Private

Your Data Stays Yours

Runs entirely local. Use your own API keys or local LLMs. No cloud dependencies. No data leaves your machine.

Open Source

Built on LangChain & LangGraph

Not a black box—LangConfig generates real LangChain code. Eject anytime. You're never locked in.

Works with your models

OpenAI Anthropic Google DeepSeek Ollama LM Studio

More from Ghost Peony

Other projects by Ghost Peony LLC

BashStats

Analytics and insights for your shell usage

BashBros

Collaborative terminal tools for teams

BashGym

Practice and sharpen your command-line skills

Build, Test, Automate, Ship

The full agent lifecycle in one place.

[Build]

Build

  • Visual Workflow Canvas – Drag, drop, and connect agents
  • Custom Agent Builder – Configure prompts & tools with AI assist
  • RAG Knowledge Base – Upload docs for semantic search
  • Multi-Model Support – OpenAI, Anthropic, Gemini, Local models
  • Multi-Agent Patterns – Supervisor and Swarm strategies for team coordination
  • Node Caching – Skip redundant execution with per-node TTL
[Test]

Test

  • Interactive Chat – Test agents with live streaming
  • Real-Time Monitoring – Watch tool selection & tokens
  • Human-in-the-Loop – Approval checkpoints with LangGraph interrupt()
  • Iterate Quickly – Tweak and rerun in seconds
[Auto]

Automate

  • Cron Scheduling – Automate recurring workflows with timezone support
  • Webhook Triggers – Fire workflows from external services with HMAC verification
  • File Watch Triggers – Trigger on file changes with glob patterns and debounce
  • Presentation Export – Generate Google Slides, PDF, or Reveal.js from results
[Ship]

Ship

  • Export to Code – Production-ready Python packages
  • Streamlit UI – Auto-generated web interface
  • JSON Sharing – Share workflows with one click
  • Local-First – Your data never leaves your machine
  • Artifact Gallery – View and bulk download generated images and files

About

Cade Russell
Founder · Ghost Peony LLC
Connect on LinkedIn cade@langconfig.com

LangConfig is built by Cade Russell at Ghost Peony LLC. The goal is straightforward: make multi-agent AI systems accessible to everyone, whether you're a developer prototyping your next idea or someone exploring agentic AI for the first time.

LangConfig exists to lower the barrier. Build visually, learn by doing, and ship when you're ready.

MIT Licensed · Actively Maintained · Open Source

Start building in 10 minutes.

Clone the repo. Build your first workflow. Ship it.

Get Started on GitHub