MCP server for ign8t project management - AI IDE integration
npm install @ign8t/mcpSeamlessly integrate ign8t project management into your AI-powered development workflow. Access your projects, documents, and tasks directly from Cursor, Claude Desktop, and other MCP-compatible AI tools.
Model Context Protocol (MCP) is an open standard that enables AI assistants to securely access external tools and data sources. With the ign8t MCP server, your AI assistant can:
- 📋 Access all your ign8t projects without switching contexts
- 📄 Read BRD, PRD, and TRD documents to understand requirements
- 📊 View and manage your backlog in real-time
- ✅ Update task statuses as you complete work
- 🔍 Search across all tasks and projects
- 🎯 Get AI-powered task recommendations
- 🧠 Access full context including related requirements and documentation
``bash`
pnpm install -g @ign8t/mcp
1. Log in to ign8t.com
2. Navigate to Settings → API Keys
3. Click "Create New API Key"
4. Give it a descriptive name (e.g., "Cursor Integration")
5. Copy the generated key (you won't see it again!)
Cursor IDE (Recommended - using npx)
Add to ~/.cursor/mcp/settings.json:
`json`
{
"mcpServers": {
"ign8t": {
"command": "npx",
"args": ["-y", "@ign8t/mcp@latest", "--api-key", "your-api-key-here"]
}
}
}
Benefits of npx approach:
- Always uses the latest version
- No global installation needed
- Cleaner configuration
Restart Cursor to activate the integration.
Claude Desktop (Recommended - using npx)
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):
`json`
{
"mcpServers": {
"ign8t": {
"command": "npx",
"args": ["-y", "@ign8t/mcp@latest", "--api-key", "your-api-key-here"]
}
}
}
For Windows: %APPDATA%\Claude\claude_desktop_config.json
Restart Claude Desktop to activate.
Alternative: Global Installation Method
If you prefer to install globally first:
`bash`
pnpm install -g @ign8t/mcp
Then configure using command line arguments (recommended) or environment variables:
Using arguments (recommended):
`json`
{
"mcpServers": {
"ign8t": {
"command": "ign8t-mcp",
"args": ["--api-key", "your-api-key-here"]
}
}
}
Using environment variables (legacy):
`json`
{
"mcpServers": {
"ign8t": {
"command": "ign8t-mcp",
"env": {
"IGN8T_API_KEY": "your-api-key-here"
}
}
}
}
Ask your AI assistant:
`
"Can you list my ign8t projects?"
If configured correctly, it will show your active projects!
Once configured, you can interact with ign8t naturally through your AI assistant:
`
"Show me all my active projects"
"Get details about the [project name] project"
"Show me the BRD for [project name]"
"Display all documents for my current project"
`
"What should I work on next?"
"Show me the backlog for [project name]"
"Search for tasks related to authentication"
"Update task [id] to in_progress"
"Mark task [id] as done with note: 'Implemented OAuth2'"
"Get full context for task [id]"
The MCP server also provides quick access to resources:
- ign8t://projects - List of all your projectsign8t://current-tasks
- - Your active tasks across all projects
| Tool | Description | Example Use |
| -------------------- | --------------------------------------------- | ------------------------------------------- |
| list_projects | List all projects with optional status filter | "Show my archived projects" |get_project
| | Get detailed project information | "Get details for project abc123" |get_documents
| | Retrieve BRD, PRD, TRD documents | "Show me the PRD for project xyz" |get_backlog
| | View epics, stories, and tasks | "Display backlog including completed items" |get_next_task
| | AI-recommended next task | "What should I work on next?" |update_task_status
| | Change task status with notes | "Set task 123 to blocked: waiting on API" |search_tasks
| | Search across tasks | "Find all tasks mentioning 'payment'" |get_task_context
| | Full task context with requirements | "Show context for task 456" |
The ign8t MCP server is designed to fit seamlessly into your development workflow:
1. Start your day: "What are my current tasks?"
2. Get context: "Show me the requirements for this task"
3. Update progress: "Mark this task as in progress"
4. Search when stuck: "Find similar tasks I've completed"
5. Complete work: "Mark task as done and get my next task"
AI assistant can't find ign8t commands
1. Verify the MCP server is installed: which ign8t-mcp
2. Check your AI tool's configuration file has the correct path
3. Restart your AI tool after configuration changes
4. Ensure your API key is valid and active
Getting "Invalid API key" errors
1. Verify your API key in the ign8t dashboard
2. Check that the key is correctly set in your configuration
3. Ensure there are no extra spaces or quotes around the key
4. Generate a new key if the current one isn't working
Can't see my projects
1. Ensure you're logged into the correct ign8t account
2. Verify the API key belongs to your account
3. Check that you have active projects in ign8t
4. Try the command "list all my projects including archived"
The MCP server connects to https://ign8t.com by default. For enterprise deployments, self-hosted instances, or local development, you can override the API endpoint:
Using npx (Recommended):
`json`
{
"mcpServers": {
"ign8t": {
"command": "npx",
"args": ["-y", "@ign8t/mcp@latest", "--api-key", "your-api-key"]
}
}
}
``
Note: Most users don't need to set a custom API URL. It's only required if you're:
- Using an enterprise or self-hosted ign8t instance
- Developing locally (e.g., http://localhost:3000)
- Using a custom domain
For contributors working on the MCP server itself:
1. Clone this repository
2. Install dependencies: pnpm install
3. Build the project:
4. Test locally:
We welcome contributions! If you'd like to contribute to the ign8t MCP server:
`bashClone the repository
git clone https://github.com/ign8t/mcp-server.git
cd mcp-server
- 📚 Documentation: docs.ign8t.com/mcp
- 🐛 Issues: GitHub Issues
- 💬 Community: Discord
- 📧 Email: support@ign8t.com
MIT © ign8t team
---
Built with ❤️ by the ign8t team to enhance your AI-powered development workflow.