ergosum-mcp

MCP (Model Context Protocol) server for ErgoSum. Exposes your workspace context to Claude Code, Cursor, Windsurf, and any MCP-compatible AI tool.

What is ergosum-mcp?

ErgoSum is a developer context platform where you write project documentation, decisions, todos, and architecture notes in a structured workspace. This MCP server lets AI assistants directly access that context without copy-pasting or manual file sharing.

When installed, your AI tool can:

• List all pages in your workspace as a tree
• Fetch full page content by slug
• Search semantically across your entire workspace
• Ask questions with AI-generated answers sourced from your docs
• Create and edit pages directly from the AI chat

Installation

npm install -g ergosum-mcp

Setup

One-command setup (recommended)

The easiest way to configure your AI tool:

# Install for Claude Code and Cursor
ergsum mcp install

# Or install for specific tools
ergsum mcp install claude
ergsum mcp install cursor
ergsum mcp install windsurf

This automatically updates the correct config file for your tool.

Manual setup

If you prefer to configure manually, add this to your tool's MCP config:

Claude Code (~/.claude/settings.json):

{
  "mcpServers": {
    "ergosum": {
      "command": "ergosum-mcp"
    }
  }
}

Cursor (~/.cursor/mcp.json):

{
  "mcpServers": {
    "ergosum": {
      "command": "ergosum-mcp"
    }
  }
}

Windsurf (~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "ergosum": {
      "command": "ergosum-mcp"
    }
  }
}

After adding the config, restart your AI tool to activate the MCP server.

Authentication

The MCP server uses the same authentication as the ErgoSum CLI. Run ergsum login first to authenticate:

npm install -g ergosum-cli
ergsum login

You can also set the ERGOSUM_TOKEN environment variable to override the stored token.

Available Tools

Once installed, your AI assistant will have access to these tools:

• list_pages — View your workspace as a tree with all page titles and slugs
• get_page — Fetch full markdown content of a page by slug
• search — Semantic search across your entire workspace
• search_in — Search within a specific project or page subtree
• ask — Ask a question and get an AI answer sourced from your workspace
• recent_pages — List recently modified pages
• workspace_status — Check workspace health and search coverage
• create_page — Create a new page with content
• edit_page — Update an existing page

Example Usage

Your AI assistant can now respond to requests like:

• "Show me the project tree"
• "What's in the architecture decisions page?"
• "Search for authentication patterns"
• "What's the current status of the API redesign?"
• "Create a new page called 'Testing Strategy' under the main project"

Links

• ErgoSum homepage
• GitHub repository
• CLI documentation