Quire

What is Quire?

An MCP server that enables AI assistants like Claude to interact with the Quire project management platform, allowing secure access to projects, tasks, and organizational data via the Quire API.

How to use Quire?

Set up a Quire OAuth app, then configure your MCP client with the environment variables QUIRE_OAUTH_CLIENT_ID and QUIRE_OAUTH_CLIENT_SECRET. Run the server via npx -y quire-mcp (stdio mode) or set MCP_TRANSPORT=http for HTTP mode. On first use, authorize via the printed OAuth URL; tokens are cached automatically.

Key features of Quire

• 60+ tools for tasks, projects, organizations, tags, and comments
• Supports stdio and HTTP transport modes
• OAuth 2.0 authentication with automatic token refresh
• Manual token mode for pre-obtained access tokens
• Cross-platform token caching (macOS, Windows, Linux)
• Compatible with Claude Desktop, Claude Code, and Cursor

Use cases of Quire

• Let an AI assistant create, update, and delete tasks in Quire projects
• Search and filter tasks across projects or organizations
• Manage project metadata, tags, and comments via natural language
• Export project data in JSON or CSV format
• Automate project management workflows through AI conversations

FAQ from Quire

What prerequisites are needed to use Quire MCP?

Node.js v22 or higher, a Quire account, and a Quire OAuth app with appropriate scopes.

How does authentication work?

The server uses OAuth 2.0. On first use, it prints an authorization URL; after you grant access, tokens are cached for future sessions and automatically refreshed.

Can I use a manual access token instead of OAuth?

Yes, set the QUIRE_ACCESS_TOKEN environment variable to skip OAuth, but manually obtained tokens expire and won't auto-refresh.

Which MCP clients are supported?

Claude Desktop (recommended), Claude Code (CLI), and Cursor – each with specific configuration examples provided.

What transport modes are available?

Two modes: stdio (default) and HTTP. STDIO spawns the server per client; HTTP runs a persistent server for multiple clients or remote deployment.