MCP Server / Overture
Overture
Overture is an open-source, locally running web interface delivered as an MCP (Model Context Protocol) server that visually maps out the execution plan of any AI coding agent as an interactive flowchart/graph before the agent begins writing code.
Installation
claude mcp add overture -- npx -y overture-mcp
npx -y overture-mcp
npm: overture-mcp
Transport
Tools (20)
Platform
Path
macOS
`~/Library/Application Support/Code/User/globalStorage/sixth.sixth-ai/settings/sixth-mcp-settings.json`
Windows
`%APPDATA%\Code\User\globalStorage\sixth.sixth-ai\settings\sixth-mcp-settings.json`
Linux
`~/.config/Code/User/globalStorage/sixth.sixth-ai/settings/sixth-mcp-settings.json`
Step
What Happens
Feature
Description
Info
What You See
Feature
Description
Control
How
Category
What It Shows
Feature
Description
Feature
Description
Feature
Description
Operation
Description
submit_plan
Submit complete plan as XML
get_approval
Wait for user approval (blocks until approved)
update_node_status
Update node status + output during execution
plan_completed
Mark plan as successfully completed
plan_failed
Mark plan as failed with error message
check_rerun
Check if user requested a node re-run
Dokumentation
https://github.com/user-attachments/assets/eeb9c4cb-c80d-42da-bf63-c0c4ecb1e5d6
🔥 The Problem
Every AI coding agent today — Cursor, Claude Code, Cline, Copilot — works the same way:
What Happens Now
- You type a prompt
- Agent immediately starts writing code
- You have zero visibility into what it's doing
- You realize it misunderstood your request
- Hundreds of lines of code need to be discarded
- You've wasted tokens, time, and patience
Text Plans Don't Help
Some agents show plans as text in chat. But text fails to show:
- Dependencies — which tasks depend on what?
- Branch points — what alternative approaches exist?
- Context requirements — which files, APIs, or secrets are needed?
- Complexity — which steps are risky?
- Progress — what's done, what's next?
✨ The Solution
Overture intercepts your AI agent's planning phase and renders it as an interactive visual flowchart — before any code is written.
The agent doesn't write a single line of code until you approve the plan.
🚀 Installation
Overture is an MCP server that works with any MCP-compatible AI coding agent. One command to install.
Claude Code
claude mcp add overture-mcp -- npx overture-mcp
Cursor
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"]
}
}
}
Cline (VS Code Extension)
Open VS Code settings → search "Cline MCP" → add:
{
"mcpServers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"]
}
}
}
GitHub Copilot
Create .vscode/mcp.json in your project root:
{
"servers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"]
}
}
}
Note: GitHub Copilot MCP requires VS Code 1.99+ and uses
serversinstead ofmcpServers.
Sixth AI (VS Code Extension)
Add to your Sixth AI MCP settings file:
| Platform | Path |
|----------|------|
| macOS | ~/Library/Application Support/Code/User/globalStorage/sixth.sixth-ai/settings/sixth-mcp-settings.json |
| Windows | %APPDATA%\Code\User\globalStorage\sixth.sixth-ai\settings\sixth-mcp-settings.json |
| Linux | ~/.config/Code/User/globalStorage/sixth.sixth-ai/settings/sixth-mcp-settings.json |
{
"mcpServers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"],
"disabled": false
}
}
}
Global Installation (Optional)
npm install -g overture-mcp
Verify It Works
Give your agent any task. Overture automatically opens at http://localhost:3031 with your plan ready for approval.
🎯 How It Works
The Flow:
| Step | What Happens | |------|--------------| | 1. Prompt | You give your agent a task: "Build a REST API with auth" | | 2. Plan | Agent generates a detailed plan with steps, branches, and requirements | | 3. Visualize | Overture renders the plan as an interactive graph | | 4. Enrich | You click nodes, attach files, select branches, fill in API keys | | 5. Approve | You click "Approve & Execute" (or press Enter) | | 6. Execute | Watch real-time as nodes pulse, complete, or fail | | 7. Control | Pause (Spacebar), resume, re-run nodes, or modify the plan mid-flight |
🛠 Features
Interactive Plan Canvas
| Feature | Description | |---------|-------------| | React Flow Canvas | Full pan, zoom, drag with smooth animations | | Streaming Parser | Plan nodes appear in real-time as the agent generates them | | Dagre Auto-Layout | Intelligent automatic positioning of nodes | | Visual Status | Pending (gray) → Active (pulsing yellow) → Completed (green) / Failed (red) | | Next Node Indicator | Blue pulse shows which node executes next | | Complexity Badges | Low (green), Medium (yellow), High (red) at a glance | | Glow Effects | Shadow glows highlight active and upcoming nodes | | Insertable Edges | Hover over edges to insert new nodes mid-plan |
Node Details Panel
Click any node to reveal its full details:
| Info | What You See | |------|--------------| | Title & Description | Full context for what this step does | | Complexity Level | Low / Medium / High with visual indicator | | Expected Output | What the step should produce | | Risks & Edge Cases | Potential issues to watch for | | Pros & Cons | For branch options, compare trade-offs |
Dynamic Fields (User Inputs)
Nodes can request input from you before execution:
| Field Type | Use Case | |------------|----------| | String | Project names, URLs, custom values | | Number | Port numbers, limits, counts | | Boolean | Yes/No toggles for options | | Select | Dropdown with predefined choices | | Secret | API keys, tokens (masked input) | | File | File paths to attach context |
Each field includes:
- Required/optional indicator
- Default values
- Help text & descriptions
- Setup instructions ("How to get an API key")
File Attachments
Attach context files to specific nodes:
- Automatic type detection — Image, code, document, or other
- Visual icons per file type
- Descriptions — add notes about why this file matters
- Delete — remove unwanted attachments
Meta Instructions
Add custom LLM instructions to any node:
"Pay special attention to error handling here" "Use the existing auth pattern from src/auth.ts" "Make sure to add tests for edge cases"
Instructions are sent to the agent right before that node executes.
Branch Detection & Selection
When the agent proposes multiple approaches:
| Feature | Description | |---------|-------------| | Auto-Detection | Branches detected from graph structure (no special markup) | | Branch Points | Nodes with multiple outgoing edges become decision points | | Selection Modal | Side-by-side comparison with pros/cons | | Complexity Comparison | See difficulty level for each option | | Visual Indicator | Selected branch highlighted on canvas | | Skip Unselected | Only your chosen path executes |
Requirements Checklist
Before you can approve, Overture shows what's needed:
- Empty required fields — counted per node
- Branch selections — which decisions are pending
- Progress indicator — visual completion tracking
- Expandable items — click to see details
- Color coding — Green (done) / Orange (pending)
The Approve button stays disabled until all requirements are met.
Execution Controls
| Control | How |
|---------|-----|
| Approve | Click button or press Enter |
| Pause | Press Spacebar mid-execution |
| Resume | Press Spacebar again |
| Re-run Node | Click failed node → "Re-run" |
| Re-run From Here | Re-execute from any node to the end |
The approval button is smart:
- 🟢 "Approve & Execute" — plan ready, requirements met
- 🟠 "Complete Requirements" — conditions unmet
- 🔵 "Executing..." — running with spinner
- 🟢 "Completed" — all done
- 🔴 "Failed" — error occurred
Structured Output
After each node executes, see rich structured output:
| Category | What It Shows | |----------|---------------| | Overview | Summary of what was accomplished | | Files Changed | Paths, lines added/removed, diffs | | Files Created | New files with line counts | | Files Deleted | Removed files | | Packages Installed | npm packages with versions | | MCP Servers Setup | Installation status (installed/configured/failed) | | Web Searches | Queries performed, results used | | Tool Calls | Which tools were used and how often | | Preview URLs | Links to deployed sites or previews | | Notes | Info, warnings, errors |
Each category is expandable — drill in without visual overload.
Output Modal
Click any completed node to see full output:
- Scrollable for long outputs
- Syntax highlighted code snippets
- Close with Escape or click outside
🏪 MCP Marketplace
Browse and attach MCP servers directly from the Overture UI.
| Feature | Description | |---------|-------------| | Built-in Marketplace | Search and discover MCP servers | | Server Details | Descriptions, authors, GitHub links, stars | | Category Browsing | Filter by use case | | Per-Node Attachment | Attach specific tools to specific steps | | Setup Instructions | See how to configure each server | | Recommended Servers | Curated list for common tasks |
When you attach an MCP server to a node, the agent gains access to those tools only for that step.
📂 Multi-Project Support
Work on multiple projects simultaneously:
| Feature | Description | |---------|-------------| | Tab Navigation | Switch between projects instantly | | Auto Registration | Projects register on first agent contact | | Isolated State | Each project has separate plans, nodes, configs | | Unread Badges | Know when other projects have updates | | Project Context | See project name, path, and agent type |
Single project? Tab bar hides automatically for a cleaner UI.
📜 Plan History & Persistence
Never lose your work:
| Feature | Description |
|---------|-------------|
| Auto-Save | Plans saved every 3 seconds |
| Local Storage | Stored in ~/.overture/history.json |
| History Browser | Slide-in panel with all past plans |
| Status Icons | Completed, failed, executing, paused |
| Progress Bars | Visual completion percentage |
| One-Click Resume | Load and continue any past plan |
| Full Context | All field values, branch selections, attachments preserved |
Resume Information
When resuming, you get complete context:
- Current node — where execution stopped
- Completed nodes — with their outputs
- Pending nodes — what's left to do
- Failed nodes — with error messages
- All configurations — field values, branches, attachments
- Timestamps — when created, when paused
✏️ Dynamic Plan Modification
Modify plans even during execution:
| Operation | Description | |-----------|-------------| | Insert Nodes | Add new steps mid-execution | | Remove Nodes | Delete steps (edges auto-reconnect) | | Replace Content | Update node title/description in-place | | Batch Operations | Multiple changes in one request |
Plan Diff View
When a plan changes, see exactly what's different:
- Added nodes — highlighted green
- Removed nodes — highlighted red
- Modified nodes — yellow with before/after comparison
- Edge changes — added/removed connections
🔌 MCP Tools (For Agent Developers)
Overture exposes 11 MCP tools for agents to interact with:
| Tool | Purpose |
|------|---------|
| submit_plan | Submit complete plan as XML |
| get_approval | Wait for user approval (blocks until approved) |
| update_node_status | Update node status + output during execution |
| plan_completed | Mark plan as successfully completed |
| plan_failed | Mark plan as failed with error message |
| check_rerun | Check if user requested a node re-run |
| check_pause | Check if user paused execution |
| get_resume_info | Get full context for resuming a paused plan |
| request_plan_update | Request incremental plan modifications |
| create_new_plan | Signal creation of a new plan |
| get_usage_instructions | Get agent-specific instructions |
🔄 Real-time WebSocket Communication
19 server-to-client message types:
connected • plan_started • node_added • edge_added • plan_ready • plan_approved • node_status_updated • plan_completed • plan_failed • plan_paused • plan_resumed • nodes_inserted • node_removed • project_registered • projects_list • history_entries • plan_loaded • resume_plan_info • plan_updated
16 client-to-server message types:
approve_plan • cancel_plan • rerun_request • pause_execution • resume_execution • insert_nodes • remove_node • register_project • subscribe_project • unsubscribe_project • get_history • load_plan • get_resume_info • save_plan • request_plan_update • create_new_plan
Relay Mode
When the WebSocket port is already in use, Overture automatically operates as a relay client, forwarding messages through the existing server. Multiple agent instances can share a single UI.
⚙️ Configuration
| Variable | Default | Description |
|----------|---------|-------------|
| OVERTURE_HTTP_PORT | 3031 | Port for the web UI |
| OVERTURE_WS_PORT | 3030 | Port for WebSocket |
| OVERTURE_AUTO_OPEN | true | Auto-open browser on start |
Setting Environment Variables
claude mcp add overture-mcp -e OVERTURE_HTTP_PORT=4000 -e OVERTURE_AUTO_OPEN=false -- npx overture-mcp
{
"mcpServers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"],
"env": {
"OVERTURE_HTTP_PORT": "4000",
"OVERTURE_WS_PORT": "4001",
"OVERTURE_AUTO_OPEN": "false"
}
}
}
}
{
"servers": {
"overture": {
"command": "npx",
"args": ["overture-mcp"],
"env": {
"OVERTURE_HTTP_PORT": "4000",
"OVERTURE_WS_PORT": "4001",
"OVERTURE_AUTO_OPEN": "false"
}
}
}
}
⌨️ Keyboard Shortcuts
| Key | Action |
|-----|--------|
| Enter | Approve plan (when ready) |
| Space | Pause / Resume execution |
| Escape | Deselect current node / Close modal |
🤝 Supported Agents
| Agent | Status | Notes | |-------|--------|-------| | Claude Code | ✅ Full | Native MCP support | | Cursor | ✅ Full | Via mcp.json config | | Cline | ✅ Full | Via VS Code settings | | GitHub Copilot | ✅ Full | VS Code 1.99+ required | | Sixth AI | ✅ Full | Built-in, zero config |
Each agent has custom-tailored prompts for optimal plan generation.
💪 Why Overture?
For Users
- Transparency — See exactly what happens before code is written
- Control — Approve, reject, or modify any plan
- Context — Attach files and instructions to the right steps
- Choice — Compare approaches and pick your path
- Visibility — Real-time progress with rich output
- Safety — Pause, resume, or re-run at any time
- History — Resume any past plan instantly
- Efficiency — No wasted tokens on rejected approaches
For AI Coding
- Trust — Makes agents predictable and controllable
- Interpretability — See AI reasoning before execution
- Universal — Works with any MCP-compatible agent
- Extensible — MCP Marketplace for tool discovery
- Open Source — MIT licensed, community-driven
- Self-Contained — No cloud dependencies
- Works Offline — Fully local execution
- Multi-Project — Manage multiple workspaces
🧑💻 Development
# Clone the repo
git clone https://github.com/SixHq/Overture.git
cd Overture
# Install dependencies
npm install
# Build all packages
npm run build
# Start MCP server (in one terminal)
cd packages/mcp-server && npm start
# Start UI dev server (in another terminal)
cd packages/ui && npm run dev
Tech Stack
| Layer | Technologies | |-------|--------------| | MCP Server | Node.js, TypeScript, Express, WebSocket (ws), SAX XML Parser | | UI | React 18, React Flow, Zustand, Framer Motion, Tailwind CSS, Vite | | Layout | Dagre (automatic graph positioning) |
🤝 Contributing
Overture is open source and we welcome contributions!
- 🐛 Report bugs at GitHub Issues
- 💡 Suggest features at GitHub Discussions
- 📖 Improve docs — PRs welcome
- 🔧 Contribute code — see CONTRIBUTING.md
All contributions are appreciated, no matter how small.
📄 License
MIT License - see LICENSE for details.