Confluence

Provides Confluence document search and content retrieval via MCP with page metadata and content access.

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "xiandan-erizo-ops-mcp": {
      "command": "uvx",
      "args": [
        "--directory",
        "/Users/hose/code/ops/ops-mcp",
        "mcp",
        "main.py"
      ],
      "env": {
        "CONFLUENCE_URL": "https://confluence.example.com",
        "CONFLUENCE_TOKEN": "your-api-token",
        "CONFLUENCE_PASSWORD": "your-password",
        "CONFLUENCE_USERNAME": "user@example.com"
      }
    }
  }
}

You can run an MCP server that provides Confluence document search and content retrieval, exposing powerful HTTP or local (stdio) interfaces for your applications to query Confluence data through the MCP framework.

How to use

Use the MCP client to search Confluence content and fetch page details. The server exposes two main tools you will use:

search_confluence lets you search by query with an optional limit and returns matching pages with excerpts.
get_confluence_page fetches detailed information about a specific Confluence page by its ID, including content, metadata, and labels.

Here are typical usage patterns you can run from your MCP client:

# Search for Confluence content
result = await mcp.use_tool("search_confluence", {
    "query": "Python",
    "limit": 5
})

# Get a Confluence page by ID
page = await mcp.use_tool("get_confluence_page", {
    "page_id": "12345"
})

# Access content via resource URIs
content = await mcp.access_resource("confluence://pages/12345")
search_results = await mcp.access_resource("confluence://search/Python")

How to install

Prerequisites you need before running the MCP server are:
- Python 3.8 or newer with pip available
- A runtime environment for your MCP server (the following steps reference uvx/uv, common MCP tooling)
- Access credentials for your Confluence instance (URL, username or API token) if you plan to search and fetch content against a real Confluence installation.

Create an environment configuration file to store Confluence access details:

CONFLUENCE_URL="your-confluence-url"
CONFLUENCE_USERNAME="your-username"
CONFLUENCE_PASSWORD="your-password"
# Or use an API token for authentication
CONFLUENCE_TOKEN="your-api-token"

Install the MCP server dependencies. You can install via editable mode or install the required packages directly.

# Install project dependencies in editable mode
uv pip install -e .

# Or install required MCP packages directly
uv pip install "mcp[cli]>=1.5.0" "atlassian-python-api>=3.41.4" "typer>=0.9.0"

Run the MCP server. Choose one of the runtime commands shown here to start the service.

# Using uvx with an explicit directory
/Users/hose/.local/bin/uvx --directory /Users/hose/code/ops/ops-mcp mcp main.py

# Or using uv run to start the MCP server
uv run --with mcp mcp run main.py

Configuration and notes

The server relies on environment configuration for Confluence access. Provide your Confluence URL and credentials in a secure way. If you plan to run in a production-like environment, consider using a dedicated environment file or secret store for the credentials and tokens.

Tools exposed by this server are designed to be consumed by your MCP client. You can search content across Confluence and retrieve page metadata and content, including the page’s title, space, version, creation and modification information, and labels.

Available tools

search_confluence

Search Confluence content by query with an optional limit; returns matching pages with excerpts and basic metadata.

get_confluence_page

Fetch detailed information for a specific Confluence page by ID, including content, metadata, creation/modification data, and labels.

Built by

VeilStrat

AI signals for GTM teams