Jira Promps MCP Server

🚀 Jira Prompts MCP Server

This repository offers a Model Context Protocol (MCP) server that can generate prompts or context information based on Jira content and provides multiple commands.

This repository draws significant inspiration from the MarkItDown MCP Server and the example Git MCP Server. Additionally, it borrows design and code elements from the mcp-atlassian repository. The main difference between this repository and mcp-atlassian is that it uses pycontribs/jira instead of atlassian-api/atlassian-python-api to interact with the Jira API and focuses on providing prompts rather than tools. This makes it particularly suitable for use with tools that only support MCP prompts, such as Zed's AI assistant.

Here is another MCP server project maintained by me: ceshine/git-prompts-mcp-server

🚀 Quick Start

📦 Installation

Manual Installation

1. Clone this repository.
2. Install dependencies: uv sync --frozen

💻 Usage Examples

As an MCP Server for Zed Editor

Add the following content to your settings.json:

"context_servers": {
  "git_prompt_mcp": {
    "command": {
      "path": "uv",
      "args": [
        "--directory",
        "/path/to/local/jira_prompts_mcp_server",
        "run",
        "jira-prompts-mcp-server",
        "https://my-company.atlassian.net", // Jira base URL
        "your_jira_account@example.com",  // Jira username
        "your_api_key"  // Jira API key (e.g., https://id.atlassian.com/manage-profile/security/api-tokens)
      ]
    },
    "settings": {}
  }
}

Commands

The server responds to the following commands:

1. jira-issue-brief <issue-key>: Get the core fields of a Jira issue. You need to provide the issue key (e.g., PROJ-123) as a parameter.
2. jira-issue-full <issue-key>: Get the core fields, comments, related issues, and subtasks. You need to provide the issue key as a parameter.

Examples:

1. /jira-issue-brief PROJ-123
2. /jira-issue-brief PROJ-155

📄 License

This project is under the MIT License. For more details, please refer to LICENSE.