Backlog MCP Server

Backlog MCP Server

Exposes Backlog project management as tools for Claude Desktop, enabling search and management of issues, comments, wiki pages, milestones, custom fields, notifications, and more.

Category
访问服务器

README

Backlog MCP Server

An MCP server that exposes Backlog project management as tools for Claude Desktop (or any MCP-compatible client). Search and manage issues, comments, wiki pages, milestones, custom fields, notifications, and more — directly from a conversation with Claude.

Features

  • 61 tools covering issues, comments, wiki, categories, issue types, milestones, custom fields, attachments, activity feeds, notifications, watchers, and stars.
  • Multi-space support — configure any number of Backlog spaces (e.g. separate client/company instances) and target them by name per tool call, without restarting or reconfiguring.
  • Real mentionsadd_comment_with_mention posts a comment that renders as a highlighted @Name mention and triggers a real Backlog notification (bell + email), matching what you get from typing @ in the Backlog UI. Plain @Name text (including from add_comment) does neither.
  • Parallel fetching for bulk operations (get_issues_by_keys, get_recent_comments_bulk, bulk_update_issue_status).
  • Read-only tools require no confirmation; every create/update/delete tool is marked in its docstring for the client to confirm with the user before calling.

Setup

1. Install dependencies

uv sync
# or
pip install -e .

Requires Python 3.11+.

2. Configure Backlog spaces

Copy .env.example to .env and fill in your space(s). Each space needs three variables sharing a common <NAME> suffix:

BACKLOG_API_KEY_<NAME>=...
BACKLOG_HOST_<NAME>=...
PROJECT_KEY_<NAME>=...

<NAME> (lowercased) becomes the source argument you pass to tools. Add as many spaces as you need by repeating the pattern with a different <NAME>. Optionally set DEFAULT_SOURCE=<name> to control which space is used when source is omitted — otherwise the first configured space is used.

Get your API key from Backlog under Personal Settings → API.

3. Register with Claude Desktop

Copy claude_desktop_config.example.json to your Claude Desktop config location, fill in the real command/args paths and credentials, and merge it into your existing mcpServers block if you already have other servers configured:

{
  "mcpServers": {
    "backlog": {
      "command": "/path/to/backlog-mcp/.venv/Scripts/python.exe",
      "args": ["/path/to/backlog-mcp/server.py"],
      "env": {
        "BACKLOG_API_KEY_SPACE1": "your_api_key_here",
        "BACKLOG_HOST_SPACE1": "your-space.backlog.com",
        "PROJECT_KEY_SPACE1": "YOUR_PROJECT_KEY"
      }
    }
  }
}

Your filled-in claude_desktop_config.json is machine-specific and contains live credentials — keep it out of version control (already covered by .gitignore).

Tools

Discovery & metadata

get_sources, get_project, get_project_statuses, get_project_users, get_issue_types, get_categories, get_milestones, get_priorities

Issues

get_issues, get_issue, get_issues_by_keys, get_parent_issue, get_child_issues, get_issue_count, create_issue, update_issue, delete_issue, bulk_update_issue_status

Comments

get_issue_comments, get_recent_comments_bulk, add_comment, add_comment_with_mention, update_comment, delete_comment

Attachments

get_issue_attachments, download_issue_attachment, get_wiki_attachments, download_wiki_attachment — images are returned inline, text files as text, other binaries as base64.

Project admin

Categories: add_category, update_category, delete_category Issue types: add_issue_type, update_issue_type, delete_issue_type Milestones: add_milestone, update_milestone, delete_milestone Custom fields: get_custom_fields, add_custom_field, update_custom_field, delete_custom_field

Wiki

get_wiki_pages, get_wiki_count, get_wiki_page, create_wiki_page, update_wiki_page, delete_wiki_page

Activity & notifications

get_space_activities, get_project_activities, get_recently_viewed_issues, get_recently_viewed_wikis, get_notifications, get_notification_count, mark_notification_read, reset_notification_count

Watchers & stars

get_issue_watchers, add_issue_watcher, delete_issue_watcher, add_star

Space / user

get_projects, get_my_info

Project layout

  • server.py — MCP tool definitions (FastMCP)
  • backlog_api.py — HTTP client, pagination, retries, response slimming (not exposed as tools)
  • config.py — builds the multi-space source registry from environment variables

License

No license specified — all rights reserved by default.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选
mcp-server-qdrant

mcp-server-qdrant

这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。

官方
精选