Jira Cloud MCP Server
Enables LLMs to interact with Jira Cloud through the official REST API v3, supporting operations like creating issues, searching with JQL, fetching issue details, and managing transitions.
README
Jira Cloud MCP Server
A production-ready, clean, and modular Model Context Protocol (MCP) server that enables LLMs to communicate directly with the official Jira Cloud REST API v3 using standard STDIO transport. Built using TypeScript strict mode and axios.
Features
- jira_get_create_metadata: Inspect projects, issue types, field requirements, and custom field IDs (
customfield_XXXXX). - jira_get_create_issue_meta_fields: Fetch the field configurations and custom fields for a specific project and issue type when creating an issue (granular and high-performance alternative).
- jira_get_project_issue_types: Fetch all issue types available for a specific project, including their IDs (used to retrieve metadata fields).
- jira_create_issue: Create new tickets, automatically converting plain text descriptions into the required Jira Cloud Atlassian Document Format (ADF). Supports optional custom fields.
- jira_get_issue: Fetch complete details for a ticket using its ID or Key.
- jira_search_jql: Query issues using Jira Query Language (JQL) with customizable limits.
- jira_find_users: Search for Jira users to retrieve their unique
accountId(required for assigning issues or setting user fields in Jira v3). - jira_get_transitions: Retrieve the workflow transitions and statuses available for a specific issue, along with their
transitionIds. - jira_transition_issue: Transition an issue to a new status (e.g., "Done", "QA") using a transition ID.
Configuration & Authentication
The server supports both Jira Data Center (PAT Bearer Token) and Jira Cloud (Basic Auth).
Jira Data Center (PAT Authentication - Recommended)
Define these environment variables:
| Environment Variable | Description | Example |
|---|---|---|
JIRA_HOST |
The root URL of your Jira Data Center instance | https://jira.yourcompany.com |
JIRA_API_TOKEN |
Your Personal Access Token (PAT) | NjkxODMy... |
[!TIP] To create a PAT in Jira Data Center, navigate to Profile > Personal Access Tokens and click Create token. Keep
JIRA_EMAILundefined (or do not set it) to trigger Bearer Token authentication.
Jira Cloud (Basic Auth)
Define these environment variables:
| Environment Variable | Description | Example |
|---|---|---|
JIRA_HOST |
The root URL of your Jira Cloud instance | https://your-domain.atlassian.net |
JIRA_EMAIL |
The email associated with your Atlassian account | user@company.com |
JIRA_API_TOKEN |
Atlassian API Token | ATATT... |
Local Setup & Development
1. Install Dependencies
npm install
2. Configure Environment
Create a .env file in the root directory for local testing:
Jira Data Center (PAT):
JIRA_HOST=https://jira.yourcompany.com
JIRA_API_TOKEN=your-personal-access-token
# Leave JIRA_EMAIL blank or omit it
Jira Cloud (Basic):
JIRA_HOST=https://your-domain.atlassian.net
JIRA_EMAIL=user@company.com
JIRA_API_TOKEN=your-api-token
3. Build the Server
Compiles the TypeScript source files into executable ESM JavaScript in the dist/ directory:
npm run build
Direct Execution (npx)
Once built or published, the server can be executed directly using Node.js:
# Run the local build
node bin/index.js
Or run via npx (if published or linked):
npx jira-mcp-server
Client Integration
To connect this MCP server to a client (e.g., Claude Desktop, Cursor, or Antigravity), use the following configuration layouts.
Claude Desktop Integration
Add the configuration snippet below to your Claude Desktop config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"jira-mcp-server": {
"command": "node",
"args": [
"/absolute/path/to/jira-mcp/bin/index.js"
],
"env": {
"JIRA_HOST": "https://your-domain.atlassian.net",
"JIRA_EMAIL": "user@company.com",
"JIRA_API_TOKEN": "your-api-token"
}
}
}
}
Cursor / Antigravity Integration
Configure a command-based MCP server in your IDE:
- Type:
command - Command:
node /absolute/path/to/jira-mcp/bin/index.js - Set the environment variables
JIRA_HOST,JIRA_EMAIL, andJIRA_API_TOKENin your system or project configuration.
Publishing to Corporate JFrog Artifactory
To distribute this package inside your organization using JFrog Artifactory as your private NPM registry:
1. Configure the Target Registry
Add a publishConfig object to your [package.json](file:///Users/yaakov.margalit/dev/jira-mcp/package.json) to redirect publishing to Artifactory:
"publishConfig": {
"registry": "https://<your-jfrog-domain>/artifactory/api/npm/<npm-repository-name>/"
}
Alternatively, create a .npmrc file in the root of the project:
registry=https://<your-jfrog-domain>/artifactory/api/npm/<npm-repository-name>/
2. Authenticate with JFrog Artifactory
Log in to your private registry using your corporate credentials:
npm login --registry=https://<your-jfrog-domain>/artifactory/api/npm/<npm-repository-name>/
3. Build & Publish
Compile the TypeScript code and upload the package:
npm run build
npm publish
4. Consume via npx
To run the server dynamically using npx from your private JFrog repository, specify the registry parameter:
npx --registry=https://<your-jfrog-domain>/artifactory/api/npm/<npm-repository-name>/ jira-mcp-server
(Optional) If you set your global npm registry configuration to point to your Artifactory NPM proxy (which resolves both local and public NPM packages):
npm config set registry https://<your-jfrog-domain>/artifactory/api/npm/<npm-repository-name>/
npx jira-mcp-server
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。