Skip to main content
Glama

๐Ÿš€ Awesome Confluence MCP Server

The most token-efficient way for AI agents to browse and analyze Confluence documentation.

MCP Python License

Topics: mcp-server python confluence-api ai-agents token-optimization markdown fastmcp


๐Ÿ“Š Token Savings at a Glance

Typical Confluence Page (2,000 words):

Format

Tokens (Avg)

Cost (GPT-4o)

Savings

Raw HTML

2,500

$0.075

-

Your Markdown

600

$0.018

76%

Save 60-80% on LLM tokens by converting Confluence pages to clean Markdown format.

A professional Model Context Protocol (MCP) server that provides token-efficient Confluence integration. Fetch, search, and convert Confluence pages to Markdown, dramatically reducing token consumption while preserving formatting and structure.

Related MCP server: mcp-confluence

๐Ÿ’ก Why Markdown Matters

The Token-Saving Advantage:

When working with LLMs, every token counts. Confluence pages in raw HTML format consume 3-5x more tokens than the same content in Markdown:

  • HTML Format: ~2,500 tokens for a typical page

  • Markdown Format: ~500-800 tokens for the same page

  • Savings: 60-80% reduction in token usage

This means:

  • โœ… Lower API costs - Fewer tokens = less money spent

  • โœ… Faster responses - Less data to process

  • โœ… Better context - Fit more pages in your context window

  • โœ… Cleaner output - Markdown is easier for LLMs to understand and work with

โœจ Features

  • ๐Ÿ” List Spaces - Browse all accessible Confluence spaces

  • ๐Ÿ”Ž Search Pages - Find pages by title or content with optional space filtering

  • ๐Ÿ“„ Fetch as Markdown - Convert any Confluence page to clean, token-efficient Markdown

  • ๐Ÿ” Secure Authentication - Uses Atlassian API tokens (never store passwords)

  • โšก Fast & Reliable - Built with FastMCP for optimal performance

  • ๐Ÿ›ก๏ธ Error Handling - Comprehensive validation and helpful error messages

๐Ÿš€ Quick Start

1. Installation

# Clone the repository
git clone https://github.com/mazhar480/awesome-confluence-mcp.git
cd awesome-confluence-mcp

# Install with pip
pip install -e .

2. Get Your Atlassian API Token

  1. Go to Atlassian API Tokens

  2. Click Create API token

  3. Give it a name (e.g., "MCP Server")

  4. Copy the token (you won't see it again!)

3. Configure Environment

# Copy the example file
cp .env.example .env

# Edit .env with your credentials
CONFLUENCE_URL=https://your-domain.atlassian.net
CONFLUENCE_EMAIL=your.email@example.com
CONFLUENCE_API_TOKEN=your_api_token_here

4. Configure Your MCP Client

For Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "confluence": {
      "command": "python",
      "args": ["-m", "server"],
      "cwd": "/path/to/awesome-confluence-mcp",
      "env": {
        "CONFLUENCE_URL": "https://your-domain.atlassian.net",
        "CONFLUENCE_EMAIL": "your.email@example.com",
        "CONFLUENCE_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

For Cline (VS Code Extension)

Add to your MCP settings:

{
  "confluence": {
    "command": "python",
    "args": ["-m", "server"],
    "cwd": "/path/to/awesome-confluence-mcp"
  }
}

Make sure your .env file is configured in the project directory.

๐Ÿ”ง Available Tools

list_spaces

List all Confluence spaces you have access to.

Parameters:

  • limit (optional): Maximum number of spaces to return (1-100, default: 25)

Example:

List my Confluence spaces

Returns:

{
  "total": 3,
  "spaces": [
    {
      "key": "DOCS",
      "name": "Documentation",
      "type": "global",
      "id": "123456",
      "url": "https://your-domain.atlassian.net/wiki/spaces/DOCS"
    }
  ]
}

search_pages

Search for pages by title or content.

Parameters:

  • query (required): Search term to match against titles and content

  • space_key (optional): Limit search to a specific space

  • limit (optional): Maximum results to return (1-50, default: 10)

Example:

Search for pages about "API documentation" in the DOCS space

Returns:

{
  "total": 5,
  "query": "API documentation",
  "space_key": "DOCS",
  "pages": [
    {
      "id": "789012",
      "title": "REST API Documentation",
      "type": "page",
      "space": {
        "key": "DOCS",
        "name": "Documentation"
      },
      "version": 12,
      "url": "https://your-domain.atlassian.net/wiki/spaces/DOCS/pages/789012"
    }
  ]
}

fetch_page_markdown

Fetch a page and convert it to Markdown format.

Parameters:

  • page_id (required): The Confluence page ID

Example:

Fetch page 789012 as markdown

Returns:

# REST API Documentation

**Space:** Documentation (DOCS)
**Version:** 12
**URL:** https://your-domain.atlassian.net/wiki/spaces/DOCS/pages/789012
**Labels:** api, rest, documentation

---

## Overview

This page documents our REST API endpoints...

### Authentication

All requests require an API token...

๐ŸŽฏ Usage Examples

Example 1: Find and Read Documentation

1. "List my Confluence spaces"
2. "Search for 'onboarding' pages in the HR space"
3. "Fetch page 123456 as markdown"

Example 2: Research a Topic

"Search for pages about 'authentication' and fetch the top 3 results as markdown"

The MCP server will:

  1. Search for relevant pages

  2. Return the search results

  3. Fetch each page and convert to Markdown

  4. Provide clean, token-efficient content for analysis

๐Ÿ”’ Security Best Practices

  • โœ… Never commit your .env file to version control

  • โœ… Use API tokens instead of passwords

  • โœ… Rotate tokens regularly

  • โœ… Limit token scope to only what's needed

  • โœ… Store tokens securely in environment variables

๐Ÿ› ๏ธ Development

# Install with dev dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Format code
black .

# Lint code
ruff check .

๐Ÿงช Testing with MCP Inspector

Want to test the tools without writing a full client? FastMCP includes a built-in MCP Inspector:

npx @modelcontextprotocol/inspector python server.py

This launches a web interface where you can:

  • โœ… Test all three tools interactively

  • โœ… See real-time request/response data

  • โœ… Validate your Confluence credentials

  • โœ… Experiment with different parameters

Perfect for: Quick testing, debugging, and demonstrating the server to others.

๐Ÿ’ฐ Sponsorship & Support

If this MCP server saves you time and tokens, consider sponsoring its development:

  • Individual Developers: GitHub Sponsors

  • Corporate Teams: I support GitHub Invoiced Billing for bulk sponsorships. Contact me for custom MCP development and enterprise support.

Why sponsor?

  • Priority bug fixes and feature requests

  • Custom tool development for your workflow

  • Direct support and consultation

  • Help maintain this free, open-source tool

๐Ÿค Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

๐Ÿ“ License

This project is licensed under the MIT License - see the LICENSE file for details.

๐Ÿ™ Acknowledgments

๐Ÿ“ž Support


Made with โค๏ธ for the MCP community

Available Tools

3 tools
fetch_page_markdownA

Fetch a Confluence page and convert it to Markdown format.

This tool retrieves page content and converts it from HTML to Markdown, reducing token usage by 60-80% compared to raw HTML while preserving formatting, links, and structure.

Inputs: page_id (required). Returns: result

list_spacesB

List all Confluence spaces accessible to the authenticated user.

Inputs: limit. Returns: result

search_pagesB

Search for Confluence pages by title or content.

Inputs: query (required), space_key, limit. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelogโ€ฆ

    TDQS

    B3.4/5.0

    Scored across 3 tools

    Disambiguation5/5

    Each tool serves a distinct purpose: fetching a specific page, listing spaces, and searching pages. No overlap exists.

    Naming Consistency5/5

    All tools use consistent snake_case naming with clear verb-noun patterns: fetch_page_markdown, list_spaces, search_pages.

    Tool Count2/5

    Only 3 tools for a Confluence API server is thin. Common operations like creating, updating, or deleting pages are missing.

    Completeness2/5

    The server covers only read operations (fetch, list, search). No create, update, or delete tools, leaving significant gaps for agent workflows.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers