Jason-MCP

Jason-MCP

Enables natural language interaction with JetBrains YouTrack, including issue management, sprint operations, team reporting, name-based assignments, and notification checks.

Category
访问服务器

README

<div align="center">

<img src="assets/header.jpg" alt="Jason-MCP" width="340">

Jason-MCP

A well-functioning YouTrack MCP for people who hate doing YouTrack busywork.

Sprints, squad reports, name-aware actions, and the notifications inbox the official one forgot.

</div>


Jason-MCP is a Model Context Protocol server that lets Claude (and any MCP client) drive JetBrains YouTrack — read and write issues, move sprints, report story points per developer/squad, assign people by name, and check what's waiting on you. It speaks plain strings, not fragile API objects, so the common operations just work.

New here / setting it up for a teammate? → docs/INTRODUCTION.md — the one-page pitch · docs/INSTALL.md — install & connect in ~10 min · skills/README.md — the one-word skills pack.

✨ What's fine-tuned here

Beyond the standard YouTrack tools, Jason-MCP adds:

  • 🗂️ Schema-aware custom fields — field types are read from the issue itself, so any State/enum/user field updates correctly (no more StateBundleElement type errors), and fields can actually be cleared (unassign everyone in one call).
  • 🏃 Sprint operations — move issues between sprints, roll sprints over, fetch sprint contents.
  • 📊 Team reporting — story points per developer and per squad.
  • 🙋 Name-aware actions — assign, set reviewer, and @-mention people by display name, not just login. Jason-MCP resolves the name for you.
  • 🔔 Notifications inbox — surface mentions and comment activity the official MCP doesn't expose.

See FINE_TUNED_FEATURES.md for the full list, and BLOCKED_OPERATIONS.md for the custom-field fixes and why they were needed.

🚀 Quick reference — common operations

These use the proven simple-string format. Pass plain strings, not nested objects.

State, priority, assignee, type, estimation

update_issue_state("DEMO-123", "In Progress")
update_issue_priority("DEMO-123", "Critical")
update_issue_assignee("DEMO-123", "john.doe")     # or a display name: "John Doe"
update_issue_type("DEMO-123", "Bug")
update_issue_estimation("DEMO-123", "4h")          # 30m · 4h · 2d · 1w · "3d 5h"

Custom fields (and clearing them)

# Set several at once
update_custom_fields("DEMO-123", {"Priority": "Critical", "Type": "Bug"})

# Clear fields — None empties them (unassign everyone, blank the squad, etc.)
update_custom_fields("DEMO-123", {
    "Assignee": None,
    "Reviewer": None,
    "Squad": None,
})
update_issue_assignee("DEMO-123", "unassigned")    # also clears the assignee

Finding, creating, linking, commenting

search_issues("bug in login")
get_project_issues("DEMO")
get_issue("DEMO-123")

create_issue(project_id="DEMO", summary="Bug in login system",
             description="Users cannot log in with special characters")

add_dependency("DEMO-123", "DEMO-124")
add_relates_link("DEMO-123", "DEMO-125")

add_comment("DEMO-123", "Fixed the login bug")
get_issue_comments("DEMO-123")

Attachments

get_issue_raw("DEMO-123")                          # raw data incl. attachments
get_attachment_content("DEMO-123", "1-456")        # download as base64
delete_attachment("DEMO-123", "1-456")             # needs permission

🛠️ Install & run (from source)

Jason-MCP runs straight from this repo over stdio — no Docker or npm package required.

git clone https://github.com/jasonsteinman/youtrack-mcp.git
cd youtrack-mcp

python -m venv .venv
source .venv/bin/activate          # Windows: .venv\Scripts\activate
pip install -r requirements.txt

Set your credentials (or put them in a .env file — it's auto-loaded):

export YOUTRACK_URL="https://your-instance.youtrack.cloud"
export YOUTRACK_API_TOKEN="perm-XXXXXXXX..."

Then run it:

python main.py                     # stdio (default, for MCP clients)
python main.py --transport sse --port 8000   # or SSE over HTTP
python main.py --version

Wire it into Claude

Add Jason-MCP to your MCP client config (e.g. Claude Desktop / Claude Code), pointing at this repo's main.py:

{
  "mcpServers": {
    "youtrack": {
      "command": "python",
      "args": ["/absolute/path/to/youtrack-mcp/main.py"],
      "env": {
        "YOUTRACK_URL": "https://your-instance.youtrack.cloud",
        "YOUTRACK_API_TOKEN": "perm-XXXXXXXX..."
      }
    }
  }
}

⚙️ Configuration

Variable Purpose
YOUTRACK_URL Your YouTrack instance URL
YOUTRACK_API_TOKEN Your YouTrack permanent API token
YOUTRACK_VERIFY_SSL SSL verification (default: true)
ENABLED_TOOLS Allowlist — enable only these tools (disables all others)
DISABLED_TOOLS Denylist — disable specific tools

Tool filtering keeps context lean. Tool names are case-insensitive and treat -/_ the same; ENABLED_TOOLS wins over DISABLED_TOOLS. Filtering happens at startup.

export ENABLED_TOOLS="get_issue,search_issues,update_issue_state"   # allowlist
export DISABLED_TOOLS="delete_issue,delete_attachment"              # denylist

🧪 Development

pip install -r requirements.txt
python -m pytest tests/unit -q          # run the unit suite

The custom-field fixes are covered by tests/unit/test_custom_field_schema_aware.py.

🐱 About the name

Jason-MCP. The kittens are just here for morale.

💬 Support

Questions, bugs, or ideas? Open an issue or reach me at maintainer@example.com.


<sub>Forked from tonyzorin/youtrack-mcp and fine-tuned. Thanks to the original author for the foundation.</sub>

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
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
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
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
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
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
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
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