M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

Atlassian Confluence MCP Server

Community
Connect AI assistants directly to your Confluence knowledge base
How to install ↓ Compare GitHub source repository ↗
Category
Collaboration #50 of 55
Stars
★ 62 Popular
Transport
stdio (local process) · Streamable HTTP
Runtime
Node.js 18+
Credentials
API key / credential required
Last commit
⚠
Tools
5
37FMRS · D

This server provides a flexible way to connect AI assistants to Confluence with comprehensive access via generic API methods. TOON format and JMESPath filtering help control token costs. However, it requires careful credential management and recommends always using filtering to avoid unexpected large responses.

Strongest · Documentation 8/20 Weakest · Reliability 6/20

Reliability
6/20
Security and permissions
8/20
Maintenance
8/20
Documentation
8/20
Setup experience
7/20
Why each score
Reliability 6/20
Source shows well-defined tools with Zod validation, but test coverage is limited, no CI workflow visible, and error handling/edge cases are under-documented. Static review cannot verify runtime; no direct contradictions found, but lack of reproducible test evidence keeps score low.
Security and permissions 8/20
Credentials are passed via env vars or config files, not hardcoded; data flows locally to Confluence API, no malicious behavior found. However, tools allow arbitrary HTTP operations without confirmation for dangerous actions, least-privilege is incomplete, but no red-line violations.
Maintenance 8/20
Repository has multiple releases (v3.x), recent updates, but no clear license, issue responsiveness unclear, dependency update status not verified. Active but governance and versioning gaps exist.
Documentation 8/20
README is rich, covering install, auth, tool params, examples, and troubleshooting, but lacks deep discussion of limitations (e.g., Cloud only) and cost (token usage). Also embeds executable instructions; treat with caution.
Setup experience 7/20
Provides multiple installation methods (npx, global, config file) and client config examples, steps are clear. Depends on Node.js, lacks platform-specific details like Windows, and setup path relies on external package manager, not fully verified.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

  • Developers who need quick access to technical documentation and API guides
  • Product managers searching for requirements, specs, and project updates
  • HR teams accessing policy documents and employee resources quickly
  • Support teams finding troubleshooting guides and knowledge base articles
  • Anyone who wants to interact with Confluence using natural language

Not for

  • Users on Confluence Server or Data Center (on-premise), as this tool supports only Confluence Cloud.
  • Legacy users expecting specific domain tools (like conf_ls_spaces), but v3.0 migrated to generic HTTP method tools.

Required permissions

  • Requires an Atlassian account with access to Confluence and the spaces you want to query.
  • Requires an API token (permissions granted automatically upon creation).
  • The server uses your own credentials and runs locally.

Risks and side effects

  • Token leakage risk: If your API token is compromised, others may access your Confluence data. Store it securely.
  • Data access scope: AI assistants can read and modify all content you have permission to access, including create, update, delete.
  • Large responses can lead to high token costs: always use jq filtering and limit parameters to avoid unexpected costs.
  • Response truncation: Responses exceeding ~40,000 characters are truncated, and the full raw data is saved to a temporary file.

Setup

Before you start

Runtime：Node.js 18+

ATLASSIAN_SITE_NAME required Your Confluence site name, e.g. mycompany from mycompany.atlassian.net.
ATLASSIAN_USER_EMAIL required Your Atlassian account login email.
ATLASSIAN_API_TOKEN requiredsecret API token created at id.atlassian.com/manage-profile/security/api-tokens; copy it immediately after creation.
Other optional settings (3)
DEBUG optional Set to true to enable detailed debug logging.
TRANSPORT_MODE optional Set to http to run the server in HTTP mode on port 3000 for testing.
PORT optional Custom port for HTTP mode, defaults to 3000.
  1. Generate an API token in your Atlassian account (https://id.atlassian.com/manage-profile/security/api-tokens).
  2. Set environment variables: ATLASSIAN_SITE_NAME (e.g., your-company), ATLASSIAN_USER_EMAIL, ATLASSIAN_API_TOKEN.
  3. For Claude Desktop, edit ~/.claude/claude_desktop_config.json and add the mcpServers configuration above.
  4. Restart Claude Desktop or configure other MCP-compatible assistants (e.g., Cursor, Continue.dev).
  5. Alternatively, run directly with npx -y @aashari/mcp-server-atlassian-confluence for CLI commands.
claude_desktop_config.json 
{
  "mcpServers": {
    "confluence": {
      "command": "npx",
      "args": ["-y", "@aashari/mcp-server-atlassian-confluence"],
      "env": {
        "ATLASSIAN_SITE_NAME": "your-company",
        "ATLASSIAN_USER_EMAIL": "[email protected]",
        "ATLASSIAN_API_TOKEN": "your_api_token"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "confluence": {
      "command": "npx",
      "args": [
        "-y",
        "@aashari/mcp-server-atlassian-confluence"
      ],
      "env": {
        "ATLASSIAN_SITE_NAME": "your-company",
        "ATLASSIAN_USER_EMAIL": "[email protected]",
        "ATLASSIAN_API_TOKEN": "your_api_token"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add confluence -e ATLASSIAN_SITE_NAME=your-company -e [email protected] -e ATLASSIAN_API_TOKEN=your_api_token -- npx -y @aashari/mcp-server-atlassian-confluence

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Add the server to your Claude Desktop config and restart; the confluence server should appear in the status bar, or run npx -y @aashari/mcp-server-atlassian-confluence get --path "/wiki/api/v2/spaces" to verify your credentials work.

Troubleshooting

  1. Authentication failed or 403: Check that the API token is still active, site name format is correct (e.g., 'mycompany' not 'https://mycompany.atlassian.net'), and verify credentials.
  2. Resource not found (404): Check path case sensitivity, use numeric IDs (not keys), and ensure you have permissions.
  3. No results when searching: Try different search terms, use proper CQL syntax, and validate in Confluence.
  4. Claude Desktop integration issues: Restart Claude, check config file path (macOS: ~/.claude/claude_desktop_config.json, Windows: %APPDATA%\Claude\claude_desktop_config.json).
  5. Enable debugging with DEBUG=true for detailed logs, stored at ~/.mcp/data/@aashari-mcp-server-atlassian-confluence.[session-id].log.

Things to try

Once connected, you can ask your AI assistant things like:

  • List all the spaces in our Confluence
  • Search for pages about API authentication
  • Show me details about the Engineering space
  • Create a new page in the DEV space titled 'API Guide'

Tools 5

conf_get read-only
Send a GET request to any Confluence API endpoint to read data.
conf_post writes
Send a POST request to create resources (e.g., pages, comments).
conf_put writes
Send a PUT request to replace resources (e.g., update a page).
conf_patch writes
Send a PATCH request to partially update resources (e.g., rename a space).
conf_delete destructive
Send a DELETE request to delete resources.

Use cases

Ask AI about documentation: "What's our API authentication process?"
Search across all spaces: "Find all pages about security best practices"
Get instant answers: "Show me the latest release notes from the Product space"
Access team knowledge: "What are our HR policies for remote work?"
Review page comments: "Show me the discussion on the architecture document"
Create and update content: "Create a new page in the DEV space"

Supported clients

Claude Desktop
Cursor AI
Continue

Listed from the project's documentation, not tested by this site.

Overview

This MCP server allows AI assistants (such as Claude, Cursor AI, etc.) to interact with Atlassian Confluence through the standard MCP interface. It provides five generic tools (conf_get, conf_post, conf_put, conf_patch, conf_delete) that can access any Confluence API endpoint, enabling AI to read, search, create, update, and delete spaces, pages, comments, and blog posts. Output defaults to TOON format for 30-60% token reduction and supports JMESPath filtering to extract only needed fields.

Similar servers

Compare with other MCP servers

Source revision 13f3b4fdf75d Data synced 2026-10-12 Read the FMRS scoring method