mcp-devtools
MCP DevTools: A suite of Model Context Protocol servers enabling AI assistants to interact with developer tools and services
claude mcp add --transport stdio dxheroes-mcp-devtools npx -y @mcp-devtools/jira \ --env JIRA_URL="https://[YOUR_WORKSPACE].atlassian.net" \ --env JIRA_API_KEY="[YOUR_API_KEY]" \ --env JIRA_API_MAIL="[YOUR_EMAIL]"
How to use
MCP DevTools provides a framework of servers that let an AI assistant interact with external tools and services via the Model Context Protocol. Currently, the package set includes integrations for Jira and Linear, enabling the AI to query tickets, create issues, search, and manage workflow tasks directly from natural language prompts. The Jira server exposes commands for fetching, reading, creating, and querying Jira tickets, while the Linear server offers similar capabilities for issues, searches, and team management. After configuring these MCP servers in Cursor IDE or your preferred client, you can invoke tasks by issuing natural language commands like get ticket SCRUM-123 or create_issue for team eng with a descriptive title and priority. The tools are designed to be embedded into conversational flows, enabling seamless interaction with your project management ecosystems.
How to install
Prerequisites:
- Node.js (recommended latest LTS) installed on your development machine
- Git to clone the repository
- pnpm as the workspace manager (recommended)
Setup steps:
-
Clone the repository: git clone https://github.com/modelcontextprotocol/mcp-devtools.git cd mcp-devtools
-
Install pnpm globally if you don’t have it: npm install -g pnpm
-
Install dependencies for the workspace: pnpm install
-
Build all packages (optional but recommended for development): pnpm build
-
Run in development mode (with auto-rebuild): pnpm dev
Notes:
- This project uses a mono-repo workspace. You can run individual packages, e.g., navigate to packages/jira or packages/linear and follow package-specific README instructions for testing.
- If you plan to publish or test locally, ensure your environment variables for Jira/Linear are set as shown in the mcp_config section.
Additional notes
Tips and notes:
- The MCP servers communicate over stdio, so debugging with the MCP Inspector can be helpful. You can run pnpm inspector to obtain a debugging URL.
- All published packages are in beta (0.x.x). Expect breaking changes across minor versions during beta.
- Environment variables required for integrations should be kept secure and not committed to source control. Use your CI/CD secret management where possible.
- When configuring in Cursor IDE, you can reuse the same command format shown in the README examples to wire up Jira and Linear integrations; ensure the appropriate environment variables are provided at runtime.
- If you don’t need a specific integration, you can disable or remove its entry from the mcp_config to reduce unnecessary resource usage.
Related MCP Servers
Mantic.sh
A structural code search engine for Al agents.
Pare
Dev tools, optimized for agents. Structured, token-efficient MCP servers for git, test runners, npm, Docker, and more.
fast -telegram
Telegram MCP Server and HTTP-MTProto bridge | Multi-user auth, intelligent search, file sending, web setup | Docker & PyPI ready
kratos
🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
mcp-ssh-orchestrator
Secure SSH access for AI agents via MCP. Execute commands across your server fleet with policy enforcement, network controls, and comprehensive audit logging.
mcp-cron
MCP server for scheduling and running shell commands and AI prompts