Dooray

You can run the Dooray MCP Server to bridge Dooray API with Claude Code, exposing six integrated tools for tasks, comments, tags, search, members, and files. This server lets you perform common Dooray operations directly through your MCP client, enabling seamless interaction and automation.

How to use

You will use an MCP client to connect to your Dooray MCP Server. Once connected, you can invoke the six tools to manage tasks, comments, and more. Use each tool by sending structured actions that describe your desired operation, such as listing tasks, creating a comment, tagging a task, or fetching file metadata. The server handles authentication, project scoping, and translating your actions into Dooray API calls.

How to install

Prerequisites you need on your machine are Python 3.8 or later, Git, and a running MCP runtime (uvx) or equivalent environment capable of executing Python modules.

# 1. Clone the repository
git clone https://github.com/tallpizza/dooray-mcp.git
cd dooray-mcp

# 2. Install dependencies (if an install script is provided, use it)
./install.sh

# 3. Create and configure environment variables
cp .env.example .env
```},

Configure the following environment variables before running the server. You will replace placeholders with your actual values.

DOORAY_API_TOKEN=your-actual-dooray-api-token
DOORAY_BASE_URL=https://api.dooray.com
DOORAY_DEFAULT_PROJECT_ID=your-default-project-id
LOG_LEVEL=INFO
```},

You will run the MCP server through the runtime command and ensure it starts successfully. The server runs a Python module that exposes the MCP interface for Claude Code.

# 4. Start the Dooray MCP server (example runtime command)
uv sync
```},

Optionally, you can register the MCP server with Claude Code using the full runtime configuration, including environment values.

# 5. Add the MCP server to Claude Code (example JSON-based config)
claude mcp add-json dooray '{
  "type": "stdio",
  "command": "uv",
  "args": ["run", "python", "-m", "dooray_mcp.server"],
  "env": {
    "DOORAY_API_TOKEN": "your-actual-dooray-api-token",
    "DOORAY_BASE_URL": "https://api.dooray.com",
    "DOORAY_DEFAULT_PROJECT_ID": "your-default-project-id"
  }
}'
```]}]},{

You can also prepare a configuration file and load it with Claude Code if you prefer not to execute the JSON command directly.

# 6. Alternative: use a config file
cp .mcp.json.example .mcp.json
# Edit .mcp.json to include your env values
claude --mcp-config .mcp.json
```]}]},{

7. Verify the connection and tools are available by listing MCP servers and trying a couple of simple tool interactions.

claude mcp list
claude --print "dooray_tasks를 사용해서 업무 목록을 조회해주세요."
```]}]} ,{

Configuration and runtime notes

The server exposes six tools, each mapped to a specific set of Dooray API endpoints. You can rely on the environment configuration to scope requests to the correct project and authentication context.

The server uses the following API base URL and authentication method:

- Base URL: https://api.dooray.com
- Authorization header: dooray-api {TOKEN}

Available tools

dooray_tasks

Task management: list, view, create, update, delete, change status, assign owner within a project.

dooray_comments

Comment management: list, create, update, delete, with support for mentions.

dooray_tags

Tag management: list, create, add or remove tags on tasks.

dooray_search

Search functionality: query tasks by text, or filter by status, tags, assignee, or date ranges.

dooray_members

User management: search by email/id, view user details, list project members.

dooray_files

File management: list task files, fetch file metadata, download file contents, access drive-based content by Content ID.