Confluence MCP Server

A Model Context Protocol (MCP) server that enables AI assistants to search and retrieve content from Confluence pages. This allows AI to understand context and design documentation before making changes.

Features

• Search Confluence Pages: Search for pages by title, content, or keywords
• Retrieve Page Content: Get full page content including body, metadata, and attachments
• List Spaces: Browse available Confluence spaces
• Get Page by ID: Retrieve specific pages by their Confluence ID
• Context-Aware Operations: AI can understand existing documentation before suggesting changes
• Weekly status drafting (full-stack): Generate weekly progress and weekly manager review drafts from Jira activity, edit in a small web UI, and copy/paste into Confluence

Problem this solves

Weekly updates are easy to forget and hard to write consistently. This project turns your Jira activity (optionally combined with a few Confluence “context” pages) into manager-ready weekly drafts with a repeatable structure: highlights, progress, risks, next steps, and asks.

Screenshot

!Weekly Progress + Manager Review web UI

Prerequisites

• Python 3.11+ recommended (works with your existing MCP setup)
• Confluence Cloud or Server instance
• Confluence API token or username/password for authentication
• Node.js 18+ (for the web UI)

Installation

Option A: Using venv (recommended)

1. Create a venv and install dependencies:

python3 -m venv .venv
.venv/bin/python -m pip install -r requirements.txt

1. Configure by creating a .env file:

cp env.example .env
# Edit .env with your Confluence credentials

Option B: Docker Compose

docker compose -f docker-compose.yml up --build

Configuration

The server requires the following environment variables:

• CONFLUENCE_URL: Your Confluence instance URL
• For Cloud: https://your-domain.atlassian.net
• For Server: https://confluence.your-domain.com
• CONFLUENCE_EMAIL: Your Confluence email/username
• CONFLUENCE_API_TOKEN: Your Confluence API token (preferred) or password

Note: The server automatically detects Confluence Cloud vs Server and uses the correct API endpoints (/wiki/rest/api/ for Cloud, /rest/api/ for Server).

Getting a Confluence API Token

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
2. Click "Create API token"
3. Copy the token and use it in your configuration

Quick Start

1. Test your connection:

python3 python/test_connection.py

1. Start the weekly draft backend API:

PYTHONPATH="python/backend:python:." .venv/bin/python -m uvicorn app.main:app --host 0.0.0.0 --port 8000

1. Start the web UI:

cd frontend
npm install
npm run dev

Then open http://localhost:5173.

1. Configure your MCP client (see docs/CONFIGURATION.md for details)

Usage

Testing the Connection

Before configuring your MCP client, test that everything works:

# Set environment variables or use .env file
export CONFLUENCE_URL="https://your-domain.atlassian.net"
export CONFLUENCE_EMAIL="your-email@example.com"
export CONFLUENCE_API_TOKEN="your-api-token"

# Run the test script
python3 python/test_connection.py

Running as MCP Server

The server implements the MCP protocol and can be used with MCP-compatible clients like Cursor, Claude Desktop, etc.

Configure in your MCP client settings:

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

Available Tools

1. search_pages: Search for Confluence pages by query

• Parameters: query (string), space_key (optional), limit (optional, default: 10)

1. get_page: Get a specific page by ID

• Parameters: page_id (string)

1. get_page_content: Get the full content of a page

• Parameters: page_id (string), expand (optional, default: "body.storage,version,space")

1. list_spaces: List all accessible Confluence spaces

• Parameters: limit (optional, default: 50)

1. search_by_title: Search pages by exact or partial title match

• Parameters: title (string), space_key (optional), limit (optional)

1. get_jira_summary: Get a summary of Jira tickets assigned to you from the last N days

• Parameters: days (optional, default: 14), save_to_file (optional, default: true)
• Returns a formatted summary grouped by status and issue type
• Automatically saves to 2026/YYYYMM/jira_summary_YYYYMMDD.md if save_to_file is true
• Uses the same Atlassian credentials as Confluence (CONFLUENCE_URL, CONFLUENCE_EMAIL, CONFLUENCE_API_TOKEN)

1. get_jira_weekly_report: Generate a high-level weekly report with business benefits

• Parameters: days (optional, default: 14), save_to_file (optional, default: true)
• Analyzes tickets to extract value propositions and formats them for stakeholders
• Categorizes into "This Week" (Done/In Progress/In Review) and "Next Week" (To Do)
• Automatically saves to 2026/YYYYMM/weekly_report_YYYYMMDD.md if save_to_file is true
• Perfect for weekly status updates to management

1. search_jira_tickets_by_email: Search for Jira tickets by a specific user's email address

• Parameters: email (required), days (optional, default: 14), search_type (optional, default: "assignee")
• Automatically looks up the user's Account ID from their email, then searches for their tickets
• search_type can be: "assignee" (tickets assigned to user), "reporter" (tickets reported by user), or "all" (both)
• Returns a formatted summary grouped by status
• Useful for understanding what a team member has been working on
• Note: Jira Cloud requires Account ID (not email) for searches. This tool handles the lookup automatically.

1. generate_weekly_drafts: Generate two markdown drafts (weekly progress + manager review)

• Parameters: week_start (optional ISO date), week_end (optional ISO date), jira_days_lookback (optional), confluence_page_ids (optional array), save_to_files (optional), save_to_db (optional)
• Saves markdown to 2026/YYYYMM/ and (optionally) stores drafts in SQLite for the web UI

Example Usage

Once configured, AI assistants can use the server like this:

User: "Search for pages about database design"
AI: [Uses search_pages tool to find relevant pages]

User: "What does the Beta Pipeline design document say?"
AI: [Uses search_pages to find the document, then get_page_content to retrieve it]

User: "Summarize my Jira tickets from the last 2 weeks"
AI: [Uses get_jira_summary to fetch and summarize tickets, saves to 2026/YYYYMM folder]

User: "Generate a weekly report with business benefits from my Jira tickets"
AI: [Uses get_jira_weekly_report to create stakeholder-friendly weekly update]

User: "What has user@example.com been working on in the last 2 weeks?"
AI: [Uses search_jira_tickets_by_email to find and summarize their tickets]

User: "Generate my weekly progress and manager review drafts for last week"
AI: [Uses generate_weekly_drafts tool, returns two markdown drafts]

Security Notes

• Never commit your .env file or API tokens to version control
• Use API tokens instead of passwords when possible
• The server respects Confluence permissions - users can only access pages they have permission to view
• If you accidentally shared a token in docs or chat, rotate it immediately in Atlassian and replace with placeholders

Troubleshooting

Authentication Errors

• Verify your API token is correct
• Check that your email matches your Confluence account
• Ensure your Confluence URL is correct (include https://)

Permission Errors

• The API token must belong to a user with appropriate Confluence permissions
• Some pages may be restricted and require specific permissions

Connection Errors

• Verify your Confluence URL is accessible
• Check network connectivity
• For Confluence Server, ensure the REST API is enabled

Development

To extend the server with additional functionality:

1. Add new tools in python/server.py following the MCP tool format
2. Implement the Confluence API calls in python/confluence_client.py
3. Update this README with new tool documentation

License

MIT