Categories
AI Agents Workflow Automation Self-Hosted AI MCP Multi-Agent Systems LLM Orchestration Visual Workflow Builder No-Code AIStars
Forks
Watchers
Developer links
ChatDev
Built around a drag-and-configure canvas of nodes and edges, ChatDev runs teams of AI agents as workflows you can watch, steer, and reuse. Nodes get a role, a model, and a system prompt from a plain form, and edges decide who speaks next. Agent nodes call OpenAI, Gemini, or any OpenAI-compatible endpoint. Human nodes pause the run and wait for your notes or attached files. Edges can branch on keywords or custom functions, extract values with regex, and loop through review-and-revise cycles guarded by loop counters and timers. Map and Tree edges fan one message out to parallel instances, and Tree mode reduces the results into a single answer. Agents can use simple, file, blackboard, or Mem0 memory, reflection thinking, built-in file, Python environment, and deep research tools, Agent Skills, and MCP servers over HTTP or stdio. Dozens of bundled workflows cover software teams, game development, data visualization, and deep research. The Launch page shows each node with a pixel-art avatar, a chat view, file uploads, and downloadable logs and artifacts. A Laboratory page runs one workflow over every row of a CSV or Excel file in parallel and tracks the success rate. The Python SDK runs YAML workflows from scripts. An illustrated tutorial and an English or Chinese interface get a first workflow running in minutes. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.
Benefits
- Build Agent Teams Without Code
- Drag nodes onto a canvas, connect them with edges, and fill in forms for model, provider, and prompt. Every workflow is also plain YAML that teams can version, review, and share.
- Watch Every Agent Work Live
- The Launch page streams node status, chat messages, and logs over WebSocket, pauses at Human nodes for your input, and lets you download logs and artifacts from each finished session.
- Bring Any Model You Already Use
- Point nodes at OpenAI, Gemini, or any OpenAI-compatible endpoint such as Ollama or LM Studio, while environment placeholders keep API keys out of the workflow files.
- Start From Ready-Made Workflow Templates
- Bundled YAML workflows cover software development, game building, data visualization, deep research, ReAct, and Reflexion patterns, so a useful multi-agent run is one click away instead of a blank canvas.
Features
- Visual Workflow Canvas
- Drag Agent, Human, Python, Subgraph, Passthrough, Literal, and loop nodes onto an infinite canvas, draw edges, and switch between the graph and its YAML configuration.
- Loops And Parallel Fan-Out
- Keyword and function conditions, regex payload extraction, loop counters and timers, majority voting, and Map or Tree edges let workflows branch, iterate, and process pieces in parallel.
- Memory, Tools, And Skills
- Attach simple, file, blackboard, or Mem0 memories, reflection thinking, built-in function tools, Agent Skills, and MCP servers over HTTP or local stdio to any agent node.
- Live Launch Console
- Pick a workflow, upload attachments, send a task prompt, and follow node states in graph or chat view, with human review steps and downloadable logs and artifacts.
- Laboratory Batch Runs
- Run one workflow across every row of a CSV or Excel input file with configurable parallel launches and log level, plus a dashboard for progress, elapsed time, and success rate.
- Python SDK Automation
- The chatdev Python package runs YAML workflows from scripts with attachments and variable overrides, returning the final node message for automation and batch jobs.