Jira MCP
@nguyenvanduocit
Jira MCP について
A Go-based MCP (Model Control Protocol) connector for Jira that enables AI assistants like Claude to interact with Atlassian Jira. This tool provides a seamless interface for AI models to perform common Jira operations including issue management, sprint planning, and workflow tra
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"script": {
"command": "/path-to/script-mcp",
"args": [
"-env",
"path-to-env-file"
]
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is Jira MCP?
Jira MCP is an opinionated Model Context Protocol server that connects AI assistants to Atlassian Jira. Built from real‑world development experience, it provides sophisticated tools for issue management, sprint tracking, development information retrieval, and workflow automation. It is designed for engineers and QC teams who need to interact with Jira through natural language in environments like Cursor or Claude Desktop.
How to use Jira MCP?
Install via Docker (ghcr.io/nguyenvanduocit/jira-mcp:latest) or as a local binary. Set three environment variables: ATLASSIAN_HOST, ATLASSIAN_EMAIL, and ATLASSIAN_TOKEN. Add the server to your MCP client’s configuration (STDIO mode). Optionally use the bundled jira-cli for terminal‑only tasks. To expose only a subset of tools, set the ENABLED_TOOLS environment variable.
Key features of Jira MCP
- Issue management: create, update, get, search, and child issues.
- Sprint management: list, get, and search active sprints.
- Development information: branches, PRs, commits linked to issues.
- Issue relationships: retrieve and link blocking/related issues.
- Version management: get and list project versions.
- Auditable history: complete issue change history and worklogs.
Use cases of Jira MCP
- Retrieve all pull requests linked to a Jira issue during code review.
- Transition issues through workflows automatically with AI assistance.
- Manage sprint backlogs and active sprints from a chat interface.
- Add worklogs and comments to issues without leaving the IDE.
- Enforce read‑only access for AI agents by limiting exposed tools.
FAQ from Jira MCP
How do I authenticate with Jira?
You need an Atlassian API token. Create one at https://id.atlassian.com/manage-profile/security/api-tokens. Provide it along with your instance host and email via the ATLASSIAN_HOST, ATLASSIAN_EMAIL, and ATLASSIAN_TOKEN environment variables.
Can I expose only a subset of tools?
Yes. Set the ENABLED_TOOLS environment variable to a comma‑separated list of tool names. Unset or empty means all tools are exposed. Unknown names are ignored and logged.
Does Jira MCP work without an MCP client?
Yes. The package includes a standalone CLI binary (jira-cli) that can be used directly in the terminal. Install it with just install-cli or go install.
What development information can I retrieve?
You can retrieve branches, pull requests, and commits linked to an issue via development tool integrations such as GitHub, GitLab, and Bitbucket.
Can I run Jira MCP in HTTP mode?
Yes. Run jira-mcp -env .env -http_port 3000 for debugging. Configure your MCP client to use http://localhost:3000/mcp as the URL.
「バージョン管理」の他のコンテンツ
Code Review Server
crazyrabbitLTCA MCP server for code reviews
MCP Atlassian
soopersetMCP server for Atlassian tools (Confluence, Jira)

Github
modelcontextprotocolModel Context Protocol Servers
Atlassian Data Center MCP
b1ffMCP servers for the Atlassian products (Bitbucket, Confluence, JIRA) of the Data Center version
MCP GitHub Project Manager
kunwarViveka mcp server to manage github project's functionality
コメント