Atlassian Confluence MCP Server

@aashari

代码仓库

基本信息

分类

记忆与知识

传输方式

stdio

作者

aashari

关于 Atlassian Confluence MCP Server

Node.js/TypeScript MCP server for Atlassian Confluence. Provides tools enabling AI systems (LLMs) to list/get spaces & pages (content formatted as Markdown) and search via CQL. Connects AI seamlessly to Confluence knowledge bases using the standard MCP interface.

#cli #typescript #ai #server #wiki #mcp #confluence #knowledge-base #atlassian #claude#external-tools#llm #anthropic #ai-integration #ai-connector #model-context-protocol #mcp-server

Atlassian Confluence MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-server-atlassian-confluence": {
      "command": "npx",
      "args": [
        "-y",
        "@aashari/mcp-server-atlassian-confluence",
        "get",
        "--path",
        "/wiki/api/v2/spaces"
      ]
    }
  }
}

Atlassian Confluence MCP Server 工具

GET any Confluence API endpoint (read data)

POST to any endpoint (create resources)

PUT to any endpoint (replace resources)

PATCH to any endpoint (partial updates)

DELETE from any endpoint (remove resources)

概览

What is Atlassian Confluence MCP Server?

A server that connects AI assistants (Claude, Cursor AI, and other MCP clients) directly to a Confluence knowledge base, enabling natural language queries to search, read, create, and update pages, spaces, comments, and blog posts.

How to use Atlassian Confluence MCP Server?

Install via npx (npx -y @aashari/mcp-server-atlassian-confluence) and set the required environment variables: ATLASSIAN_SITE_NAME, ATLASSIAN_USER_EMAIL, and ATLASSIAN_API_TOKEN. Configure the server in Claude Desktop’s config file (~/.claude/claude_desktop_config.json) or in another MCP‑compatible client. Alternatively, use the CLI commands (get, post, put, patch, delete) directly from the terminal.

Key features of Atlassian Confluence MCP Server

Five generic tools (conf_get, conf_post, conf_put, conf_patch, conf_delete) for any Confluence API endpoint
Optional JMESPath (jq) filtering to extract only the needed fields and reduce token costs
TOON (Token‑Oriented Object Notation) output format, which uses 30–60% fewer tokens than JSON
Automatic truncation of responses larger than ~40,000 characters, with a notice
Supports all common Confluence API paths: spaces, pages, comments, blog posts, and search

Use cases of Atlassian Confluence MCP Server

Ask AI about your documentation (e.g., “What’s our API authentication process?”)
Search across all Confluence spaces using natural language or CQL queries
Retrieve and display page content, labels, and child pages
Create new pages and add comments to existing documents
Update or delete pages, spaces, and comments via the API

常见问题

How do I authenticate with Confluence?

Set the environment variables `ATLASSIAN_SITE_NAME` (your company’s subdomain), `ATLASSIAN_USER_EMAIL`, and `ATLASSIAN_API_TOKEN`. Generate an API token from your Atlassian account settings.

What tools are available?

The server provides five tools that mirror standard HTTP methods: `conf_get`, `conf_post`, `conf_put`, `conf_patch`, and `conf_delete`. Each accepts a `path` (the API endpoint), optional `queryParams`, a `jq` filter expression, and an `outputFormat` (TOON or JSON).

How can I reduce token costs?

By default responses are returned in the TOON format, which is more token‑efficient than JSON. You can further reduce costs by using the `jq` parameter to filter the response to only the fields you need.

记忆与知识分类下的更多 MCP 服务器

server-noonien - shared memory MCP

sequico

Shared, serverless, conflict-free memory for AI agents: a multi-machine drop-in for the MCP memory knowledge graph.

Groundwork

Rarefied-Earth

Groundwork: company memory for your AI tools. Docs public. Source closed. Public proof needs no account.

Leviathan Systems Public Content MCP

Leviathan Systems

Search published GPU deployment, rack integration and structured cabling resources; read canonical documents and Leviathan Systems company service scope. Anonymous, read-only Streamable HTTP MCP. No credentials required.

ctxfile

Local-first MCP server that snapshots your project's working state into one context object. Open-core, privacy-first.

Stele

Stele-Dev

Shared memory for AI coding agents. Claude Code, Cursor and Codex read your project's decisions, lessons and tasks before they act.