# nguyenvanduocit/jira-mcp [Health: Active]

**Category:** 🎧 Support & Service Management  
**Repository:** https://github.com/nguyenvanduocit/jira-mcp  
**GitHub Stars:** 97  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/nguyenvanduocit-jira-mcp

## Description
A Go-based MCP 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 transitions.

## Tools
Capabilities this server exposes over MCP:

- **jira_get_issue** — Retrieve detailed information about a specific issue including status, assignee, description, subtasks, and available transitions
- **jira_create_issue** — Create a new issue with specified details (returns key, ID, and URL)
- **jira_create_child_issue** — Create a child issue (sub-task) linked to a parent issue
- **jira_update_issue** — Modify an existing issue's details (supports partial updates)
- **jira_list_issue_types** — List all available issue types in a project with their IDs, names, and descriptions
- **jira_search_issue** — Search for issues using JQL (Jira Query Language) with customizable fields and expand options
- **jira_list_sprints** — List all active and future sprints for a specific board or project
- **jira_get_sprint** — Retrieve detailed information about a specific sprint by its ID
- **jira_get_active_sprint** — Get the currently active sprint for a given board or project
- **jira_search_sprint_by_name** — Search for sprints by name with exact or partial matching
- **jira_list_statuses** — Retrieve all available issue status IDs and their names for a project
- **jira_transition_issue** — Transition an issue through its workflow using a valid transition ID
- **jira_add_comment** — Add a comment to an issue (uses Atlassian Document Format)
- **jira_get_comments** — Retrieve all comments from an issue
- **jira_add_worklog** — Add a worklog entry to track time spent on an issue
- **jira_get_issue_history** — Retrieve the complete change history of an issue
- **jira_get_related_issues** — Retrieve issues that have a relationship (blocks, is blocked by, relates to, etc.)
- **jira_link_issues** — Create a link between two issues, defining their relationship
- **jira_get_version** — Retrieve detailed information about a specific project version
- **jira_list_project_versions** — List all versions in a project with their details
- **jira_get_development_information** — Retrieve branches, pull requests, and commits linked to an issue via development tool integrations (GitHub, GitLab, Bitbucket)

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `docker` (confidence: high):

```json
"mcpServers": {
  "jira-mcp": {
    "command": "docker",
    "args": ["run","ghcr.io/nguyenvanduocit/jira-mcp:latest"],
    "env": {
      "ATLASSIAN_HOST": "",
      "ATLASSIAN_EMAIL": "",
      "ATLASSIAN_TOKEN": "",
      "ENABLED_TOOLS": ""
    }
  }
}
```

**Requires environment variables:** `ATLASSIAN_HOST`, `ATLASSIAN_EMAIL`, `ATLASSIAN_TOKEN`, `ENABLED_TOOLS` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What nguyenvanduocit/jira-mcp MCP server does

The nguyenvanduocit/jira-mcp MCP server connects an MCP-compatible AI client to Atlassian Jira. It covers common engineering and project-tracking operations rather than only issue lookup. Agents can retrieve, create, update, and search issues; create subtasks; inspect issue types and statuses; and transition issues through Jira workflows.

Sprint support includes listing active and future sprints, retrieving a sprint by ID, finding the active sprint, and searching sprint names. The server also handles comments, worklogs, issue history, linked issue relationships, project versions, and development data associated with an issue. Development information can include branches, pull requests, and commits from connected GitHub, GitLab, or Bitbucket integrations.

## How it works

The server is written in Go and communicates with MCP clients through STDIO, so a local binary or Docker container does not need to expose a network port for normal use. An optional HTTP mode is available for debugging and can serve the MCP endpoint locally.

Jira access uses the Atlassian host, Atlassian email, and an API token. Each operation is exposed as a named MCP tool with typed inputs. Search uses Jira Query Language, while issue comments use Atlassian Document Format. Workflow transitions require a valid Jira transition ID.

The nguyenvanduocit/jira-mcp MCP server registers all Jira tools by default. The `ENABLED_TOOLS` environment variable accepts a comma-separated allowlist, allowing an administrator to expose only selected tools. Unset or empty configuration keeps the default of exposing all tools; unknown names are ignored and logged at startup.

## Setup and configuration

Create an Atlassian API token through the Atlassian account security settings, then configure:

- `ATLASSIAN_HOST`: the Jira instance URL
- `ATLASSIAN_EMAIL`: the Atlassian account email
- `ATLASSIAN_TOKEN`: the Jira API token
- `ENABLED_TOOLS`: optional comma-separated tool allowlist

The README documents Docker and local-binary configurations for Cursor, with the credentials passed as environment variables. It also describes a local `.env` file for running the binary and an HTTP command using `-env` and `-http_port` flags. Claude Desktop is named as an installation target, but the supplied material does not provide a dedicated Claude configuration example.

## Tools and capabilities

Available tool groups include:

- Issue retrieval, creation, partial updates, subtasks, and JQL search
- Issue types, statuses, transitions, comments, worklogs, and history
- Related-issue lookup and issue linking
- Sprint listing, lookup, active-sprint retrieval, and name search
- Project-version listing and version details
- Branch, pull-request, and commit information linked to an issue

The repository also includes a separate `jira-cli` binary for terminal use, but that CLI is distinct from the MCP interface.

## Limitations and notes

The server requires access to a Jira instance and valid Atlassian credentials. Jira permissions and project configuration determine which issues, sprints, transitions, comments, versions, and development links are available. Mutating operations remain exposed unless the tool allowlist is configured to exclude them. The provided material does not specify support for OAuth, Jira Server/Data Center compatibility, or the exact permission requirements for each operation.

The nguyenvanduocit/jira-mcp MCP server is licensed under MIT.

_Full upstream README: https://allmcps.com/mcp/nguyenvanduocit-jira-mcp/readme_