EchoFlow Code
English · 简体中文
EchoFlow Code is a local-first desktop workspace for Claude Code and model providers. Run coding sessions, inspect every diff, coordinate Agent Teams and Dynamic Workflows, and keep control of every tool call and model connection.
| Inspect every change | Coordinate complex work | Keep control of your setup |
|---|
| Tool calls, workspace diffs, and undo stay visible. | Subagents, Agent Teams, and Dynamic Workflow make parallel work readable. | Connect official OAuth, vendor APIs, local models, or custom compatible endpoints. |
Desktop Preview
Install the Desktop App
- Download the macOS, Windows, or Linux installer from Releases.
- Configure a model provider, API key, and default model in Desktop Settings.
- Public macOS releases require signing and notarization. Draft or unsigned builds may require one-time manual approval. Unsigned Windows installers can show SmartScreen; choose “More info” then “Run anyway.” See the installation guide.
Run the CLI from Source
bun install
cp .env.example .env
./bin/echoflow-code
See environment variables and CLI setup for configuration details.
Why EchoFlow Code
EchoFlow Code turns a coding-agent runtime into a visible, reviewable workspace rather than another opaque chat window.
- Coordinate real multi-agent work: the Agent Teams workbench shows members, a shared task list, communication, dependency lanes, and completed-session history. Dynamic Workflow runs parallel or pipeline orchestration with phase progress, interruption, and recovery.
- Keep every change inspectable: sessions expose permission requests, tool calls, workspace diffs, browser previews, model traces, diagnostics, and an undo path instead of hiding agent execution behind a single response.
- Use the setup you trust: connect official Claude, ChatGPT, and Grok sign-in flows; vendor APIs; LM Studio or Ollama; and compatible custom endpoints. Provider credentials, model selection, and permissions remain under your control.
- Continue beyond one desktop: keep the same session moving through H5, Android, and messaging integrations while the desktop remains the source of truth for project context and approval.
- Recover with evidence: local activity, diagnostics exports, upgrade-safe persistence, and controlled repair paths help investigate startup, provider, or session failures without exposing secrets.
Desktop Highlights
- Multi-session workspace: tabs, project switching, terminal entry, and session history in one place, with a resizable sidebar.
- Global search: press Cmd+K to search across every session and jump to the match.
- Branch / Worktree launch: choose a repository branch and decide whether to use the current working tree or an isolated Worktree.
- Review edits file by file: the workspace lists this turn's changes; open any file for a syntax-highlighted diff, or undo the whole turn.
- Built-in browser preview: the page your agent just edited renders right inside the app, cookies and login state included.
- Five permission modes: from "ask every time" to "skip permissions" — risky commands, tool calls, and follow-up questions are all approved in the GUI.
- Bring your own model: sign in to Claude, ChatGPT, or Grok; use presets for DeepSeek, Kimi, Zhipu GLM and others; or point it at LM Studio and Ollama running locally.
- Image generation: generate and edit images right in the chat — sign in with ChatGPT or Grok for instant use, or plug in any OpenAI-compatible Images API.
- Visual MCP manager: add and edit MCP servers in a GUI — STDIO / Streamable HTTP / SSE, with project, shared, or global scope.
- Six colour themes: white, paper, warm classic, celadon, ink night, and ink blue — optionally following your system's light/dark setting.
- Skill marketplace: discover, preview, and install third-party skills from ClawHub / SkillHub, with source and safety status shown up front.
- Session activity panel: track task progress, background tasks, SubAgents, and sources in one side panel.
- Visual SubAgent manager: create and tune SubAgents in a GUI — model, tools, and permission mode.
- Agent Teams workbench: visualize multi-agent collaboration in the GUI — members, tasks, a communication feed, and a dependency-lane canvas.
- Dynamic Workflow orchestration: the model writes and runs orchestration scripts on the fly, driving subagents concurrently or in pipelines, with phase views, interrupts, and resume.
- Model trace: every model request is logged locally with status and timing — search and filter to diagnose stuck or failed calls.
- Computer Use: let the agent take screenshots, click, type, and control desktop apps after authorization. On macOS the native runtime works in the background without taking over your real mouse and keyboard.
- : Dada, Huhu, Bubu, and Huihui change what they do with the task at hand — or raise one of your own (off by default).
Documentation
Full documentation: https://code.echoflow.cn
Feedback and discussions
Use Discussions for questions, ideas, and usage notes. Use Issues for reproducible bugs. Read SUPPORT.md before sharing logs, and use SECURITY.md for vulnerabilities.
Stack
| Category | Technology |
|---|
| Language | TypeScript |
| Desktop | Electron |
| Desktop UI | React + Vite |
| Local runtime | Bun |
| Terminal UI | React + Ink |
| CLI parsing | Commander.js |
| API | Anthropic SDK |
| Protocols | MCP, LSP |
Acknowledgements
- React: UI ecosystem and component model.
- Electron: cross-platform desktop capabilities.
- cc-switch: provider configuration references.
