@erold/mcp-server

MCP server for AI-powered project management

Let Claude, Cursor, and other AI assistants manage your Erold projects directly.

Website · Documentation · Report Bug

What is This?

This MCP (Model Context Protocol) server connects AI assistants to Erold, enabling them to:

• Create and manage tasks
• Track project progress
• Access your knowledge base
• Update status as they work

No more copy-pasting between your AI assistant and project management tools.

How It Works

┌─────────────────┐        ┌──────────────────┐        ┌─────────────────┐
│   AI Assistant  │  MCP   │  @erold/mcp-server │  API  │   Erold Cloud   │
│  (Claude, etc.) │───────►│    (This package) │───────►│                 │
└─────────────────┘        └──────────────────┘        └─────────────────┘

Your AI can now say things like:

"Create a task for implementing OAuth in the Backend project"

And it actually happens.

Quick Start

1. Get Your API Key

1. Log in to app.erold.dev
2. Go to Settings → API Keys
3. Create a new key (starts with erold_)

2. Configure Your AI Assistant

Claude Code (Recommended)

Run this command in your terminal:

claude mcp add-json erold '{"command":"npx","args":["-y","@erold/mcp-server@latest"],"env":{"EROLD_API_KEY":"YOUR_API_KEY","EROLD_TENANT":"YOUR_TENANT_ID"}}' --scope user

Verify the connection:

claude mcp list

You should see: erold: npx -y @erold/mcp-server@latest - ✓ Connected

Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "erold": {
      "command": "npx",
      "args": ["-y", "@erold/mcp-server@latest"],
      "env": {
        "EROLD_API_KEY": "erold_your_api_key",
        "EROLD_TENANT": "your-tenant-id"
      }
    }
  }
}

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "erold": {
      "command": "npx",
      "args": ["-y", "@erold/mcp-server@latest"],
      "env": {
        "EROLD_API_KEY": "erold_your_api_key",
        "EROLD_TENANT": "your-tenant-id"
      }
    }
  }
}

Important: Use actual values for environment variables. Do NOT use ${EROLD_API_KEY} syntax - Claude apps don't expand shell variables.

3. Start Using It

Ask your AI assistant:

• "What tasks are assigned to me?"
• "Create a high-priority task for fixing the login bug"
• "Mark task TASK-123 as complete"
• "What's the status of the Backend API project?"

Claude Code Plugin

For the full Erold experience with skills, agents, and hooks, install the dedicated Claude Code plugin:

claude plugin install erold-dev/claude-plugin --scope user

Important: Configure your credentials in ~/.claude/mcp.json (not the plugin directory) so they survive plugin updates. See the plugin setup guide.

Skills (Slash Commands)

| Skill | Description | |-------|-------------| | /erold:context | Load workspace context (UNDERSTAND phase) | | /erold:plan | Create tasks from requirements (PLAN phase) | | /erold:execute | Work on a task with full context (EXECUTE phase) | | /erold:learn | Save learnings to knowledge base (LEARN phase) | | /erold:guidelines | Fetch coding guidelines for technologies | | /erold:task | Quick task operations (list, start, complete) | | /erold:search | Search tasks and knowledge | | /erold:status | Dashboard and progress overview |

Agents

| Agent | Description | |-------|-------------| | erold-workflow | Enforces the 4-phase methodology | | erold-reviewer | Reviews code against guidelines | | erold-learner | Extracts patterns from completed work |

Commands

| Command | Description | |---------|-------------| | /erold:init | Initialize Erold in current project | | /erold:sync | Sync local work with Erold PM | | /erold:report | Generate progress report |

Hooks

The plugin includes hooks for workflow enforcement:

• Session start: Auto-load context
• Pre-edit: Check for active task
• Post-edit: Log activity
• Pre-commit: Code review
• Task complete: Suggest learnings

The Erold Methodology

┌─────────────┐    ┌─────────────┐    ┌─────────────┐    ┌─────────────┐
│ UNDERSTAND  │ → │    PLAN     │ → │   EXECUTE   │ → │    LEARN    │
│             │    │             │    │             │    │             │
│ Load context│    │ Create tasks│    │ Implement   │    │ Capture     │
│ Fetch info  │    │ Break down  │    │ Track work  │    │ patterns    │
└─────────────┘    └─────────────┘    └─────────────┘    └─────────────┘

Available Tools

Task Management

| Tool | Description | |------|-------------| | list_tasks | List tasks with filters (project, status, priority, assignee) | | get_task | Get detailed task information | | create_task | Create a new task | | update_task | Update task fields | | start_task | Start working on a task | | complete_task | Mark task as complete | | block_task | Mark task as blocked with reason | | search_tasks | Search tasks by keyword | | add_task_comment | Add a comment to a task | | get_task_comments | Get all comments on a task | | get_blocked_tasks | Get all blocked tasks |

Project Management

| Tool | Description | |------|-------------| | list_projects | List all projects | | get_project | Get project details | | create_project | Create a new project | | update_project | Update project fields | | get_project_stats | Get project statistics | | get_project_tasks | Get tasks in a project |

Knowledge Base

| Tool | Description | |------|-------------| | search_knowledge | Search knowledge base | | get_knowledge | Get article content | | list_knowledge | List articles by category | | create_knowledge | Create new article | | update_knowledge | Update existing article |

Context & Insights

| Tool | Description | |------|-------------| | get_context | Get AI-ready workspace context | | get_dashboard | Get dashboard overview | | get_stats | Get workspace statistics | | get_workload | Get team workload distribution | | list_members | List team members |

Guidelines

| Tool | Description | |------|-------------| | get_guidelines | Fetch coding guidelines by category | | list_guidelines | List all available guidelines | | search_guidelines | Search across all guidelines |

Example Conversations

Starting a Coding Session

You: "What should I work on today?"

Claude: Uses get_context and list_tasks

"You have 3 high-priority tasks:

1. TASK-45: Fix authentication timeout (Backend API)
2. TASK-52: Add input validation (Backend API)
3. TASK-58: Update API documentation (Documentation)

I'd suggest starting with TASK-45 since it's blocking other work."

You: "Start working on task 45"

Claude: Uses start_task

"Started TASK-45. I've updated the status to 'In Progress' and assigned it to you."

Creating Tasks

You: "Break down the OAuth implementation into tasks"

Claude: Uses create_task multiple times

"I've created the following tasks in the Backend API project:

1. TASK-78: Set up OAuth provider configuration
2. TASK-79: Implement authorization endpoint
3. TASK-80: Add token refresh logic
4. TASK-81: Write OAuth integration tests

All set to high priority."

End of Day

You: "Update my progress - I finished the auth timeout fix and made 50% progress on validation"

Claude: Uses complete_task and update_task

"Done! TASK-45 is now complete, and I've updated TASK-52 to 50% progress."

Configuration

Environment Variables

| Variable | Required | Description | |----------|----------|-------------| | EROLD_API_KEY | Yes | Your Erold API key (starts with erold_) | | EROLD_TENANT | Yes | Your tenant ID or slug | | EROLD_API_URL | No | Custom API URL (default: production) |

Finding Your Tenant ID

1. Log in to app.erold.dev
2. Go to Settings → Workspace
3. Copy your tenant ID or slug

Supported AI Assistants

Works with any MCP-compatible client:

• Claude Desktop — Anthropic's desktop app
• Claude Code — VS Code extension
• Cursor — AI-first code editor
• Windsurf — Codeium's AI IDE
• Cline — VS Code AI assistant
• Continue.dev — Open source AI assistant
• Custom agents — Any MCP-compatible implementation

Development

# Clone the repo
git clone https://github.com/erold-dev/mcp-server.git
cd mcp-server

# Install dependencies
npm install

# Build
npm run build

# Run tests
npm test

# Run locally
npm start

Testing with MCP Inspector

# Set environment variables
export EROLD_API_KEY="erold_your_key"
export EROLD_TENANT="your-tenant"

# Run inspector
npm run inspect

Troubleshooting

"API key invalid"

• Verify your key starts with erold_
• Check the key hasn't expired in Settings → API Keys
• Ensure you're using the correct tenant ID

"Connection refused"

• Check your internet connection
• Verify the API URL if using a custom one
• Try restarting your AI assistant

Tools not appearing

• Restart your AI assistant after config changes
• Verify the config file path is correct
• Check for JSON syntax errors in config

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Related

• @erold/cli — Command-line interface
• Erold Web App — Full web interface
• Documentation — Complete documentation
• MCP Specification — Model Context Protocol

License

MIT © Erold