jira-pilot
AI-powered Jira CLI and MCP server for humans and agents manage issues, sprints, boards with interactive wizards, multi-provider AI (OpenAI/Gemini/Anthropic), and an 14-tool MCP server for AI assistants
claude mcp add --transport stdio aarul5-jira-pilot npx -y jira-pilot
How to use
Jira Pilot is a dual-purpose tool: it provides a human-friendly CLI for Jira workflows and also functions as an MCP (Model Context Protocol) server exposing 14 MCP-compatible tools that let AI agents interact with a Jira instance safely. As an MCP server, it communicates via standard MCP stdio transport, returning clean, structured JSON responses that clients can parse efficiently. Humans can also leverage Jira Pilot as a powerful interactive CLI for issue management, sprints, boards, and AI-assisted coding tasks. To use the MCP capabilities, launch the server (via your preferred MCP runtime) and connect an MCP client that can issue the listed tools such as list_issues, get_issue, create_issue, update_issue, transition_issue, assign_issue, add_comment, add_worklog, create_subtask, add_attachment, search_users, myself, list_projects, and list_sprints. The server ensures safe access patterns by returning well-formed JSON objects suitable for downstream agents and automation pipelines.
How to install
Prerequisites:
- Node.js 20.0.0 or higher
- npm (comes with Node.js)
Installation steps:
- Install the Jira Pilot package globally (recommended for direct CLI usage):
npm install -g jira-pilot
- Verify installation:
jira --version
- Start the MCP-enabled server using npx (as configured in mcp_config):
npx -y jira-pilot
- Connect an MCP client to the server and invoke tools like list_issues, get_issue, or create_issue. If you prefer a local, project-scoped setup, you can also install Jira Pilot in a project and run it from your project directory instead of globally:
npm install jira-pilot --save-dev
node node_modules/jira-pilot/dist/server.js
Prerequisites recap:
- Node.js 20.x or newer
- npm or yarn for dependency management
- Access to a Jira instance (cloud or server) with API token or credentials configured via jira config setup
Additional notes
Tips and common considerations:
- The MCP server returns structured JSON; use it with MCP clients or automation scripts to minimize token usage.
- Ensure Jira credentials and API tokens are configured prior to first use (jira config setup).
- If you run into authentication issues, verify the API token permissions and Jira site URL.
- When running as a background service, consider capturing logs to diagnose issues with tool availability or Jira API responses.
- The npm package name is jira-pilot; the MCP server name in the config should reflect your deployment (e.g., jira-pilot).
- If you upgrade Node.js, re-install the package to ensure compatibility with the MCP runtime.
- For production deployments, consider containerizing the server (Docker) and binding environment variables for credentials securely.
Related MCP Servers
wcgw
Shell and coding agent on mcp clients
claude-code-open
Open source AI coding platform with Web IDE, multi-agent system, 37+ tools, MCP protocol. MIT licensed.
mcp -atlassian-jira
Node.js/TypeScript MCP server for Atlassian Jira. Equips AI systems (LLMs) with tools to list/get projects, search/get issues (using JQL/ID), and view dev info (commits, PRs). Connects AI capabilities directly into Jira project management and issue tracking workflows.
skene-growth
Product-Led Growth (PLG) analysis toolkit that detects tech stacks, plans growth loops and builds the loop iteratively.
mcpx
Token-efficient MCP client: TypeScript schemas instead of JSON, LLM-friendly syntax, batch calls, TOON output. Built for Claude/GPT automations.
local-skills
Universal MCP server enabling any LLM or AI agent to utilize expert skills from your local filesystem. Reduces context consumption through lazy loading. Works with Claude, Cline, and any MCP-compatible client.