Use the following guide to quickly run a Dart MCP tool and chatbot in a disposable cloud Room for testing.

​
Required Environment Variables

  • DART_HOST: (e.g. https://app.itsdart.com)
  • DART_TOKEN: (e.g. dsa_... — your Dart API token)

​
Steps

  1. Install Meshagent CLI and dependencies
    • pip install "meshagent[all]"
  2. Sign Up & Authenticate to Meshagent Cloud
  3. Start a Test MCP Agent Service in a Temporary Room 
    meshagent service test \
  --room=test \
  --role=agent \
  --image=meshagent/mcp-dart:latest \
  --env MESHAGENT_PORT=8001 \
  --env DART_HOST=https://app.itsdart.com \
  --env DART_TOKEN=dsa_... \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-dart-service-test
    • Starts a Meshagent Room named test with the Dart MCP agent available.
    • Note: Rooms automatically close (removing all services/chatbots) if inactive.
  4. Start a Chatbot Using the Dart Toolkit in the Same Room 
    meshagent chatbot join \
  --room=test \
  --agent-name=mcp-dart \
  --name=mcp-dart \
  --toolkit=mcp-dart
    • A chatbot using the Dart toolkit will appear in the Room and be able to call the above agent/services.
    • Multiple toolkits can join the same Room with the same agent.
    • A direct Room link will appear in the command output.
  5. Try It Out in the Browser
    • Visit the Room link provided and send a message to the agent/chatbot to interact with Dart MCP tools.
Environment Variables Used:
  • DART_HOST=https://app.itsdart.com
  • DART_TOKEN=dsa_...

​
Project-Level Deployment (Persistent MCP Tools in All Rooms)

For production or always-on scenarios, deploy MCP tools and chatbots as persistent services at the project level. This ensures they appear instantly every time a Room is created in the project, with no need for local startup or manual Room joins.

​
Persistent Deployments

  1. Deploy MCP Agent Service Project-Wide 
    meshagent service create \
  --role=agent \
  --image=meshagent/mcp-dart:latest \
  --env MESHAGENT_PORT=8001 \
  --env DART_HOST=https://app.itsdart.com \
  --env DART_TOKEN=dsa_... \
  --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
  --name=mcp-dart-service
  2. Deploy Dart-Toolkit Chatbot Service Project-Wide 
    meshagent service create \
  --image="meshagent/cli:latest" \
  --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-dart-chatbot" \
  --env="MESHAGENT_PORT=9001" \
  --name="mcp-dart-chatbot-service" \
  --command="meshagent chatbot service --agent-name=mcp-dart-chatbot --toolkit=mcp-dart"
  • Result: Each time a Room is created in this project, both the MCP server tools and the Dart chatbot will auto-join—ensuring a consistent, managed workspace.
  • Environment Variables Used:
    • DART_HOST=https://app.itsdart.com
    • DART_TOKEN=dsa_...

​
Tools Available

Dart MCP Server exposes the following tools, available automatically when running in Meshagent Rooms:
ToolShort Description
add_task_commentAdd a comment to an existing task without modifying the task description.
create_docCreate a new doc in Dart.
create_taskCreate a new task in Dart.
delete_docMove an existing doc to the trash, where it can be recovered if needed.
delete_taskMove an existing task to the trash, where it can be recovered if needed.
get_configGet information about the user’s space, including available endpoint values.
get_dartboardRetrieve a dartboard by its ID.
get_docRetrieve a doc by its ID.
get_folderRetrieve a folder by its ID.
get_taskRetrieve a task by its ID.
get_viewRetrieve a view by its ID.
list_docsList docs from Dart with optional filtering.
list_task_commentsList comments from Dart with optional filtering.
list_tasksList tasks from Dart with optional filtering.
update_docUpdate an existing doc.
update_taskUpdate an existing task.

​
Tool Details

​
add_task_comment

Add a comment to an existing task without modifying the description. Supports markdown.
  • taskId (string): The 12-character alphanumeric ID of the task
  • text (string): Comment content (markdown allowed)

​
create_doc

Create a new Dart doc.
  • title (string): The doc title (required)
  • folder (string, optional): Folder title
  • text (string, optional): doc content (markdown allowed)

​
create_task

Create a new task with rich metadata.
  • Parameters: title, assignee[s], customProperties, dartboard, description, dueAt, parentId, priority, size, startAt, status, tags, type

​
delete_doc

Move a doc to trash (recoverable).
  • id (string): Doc ID

​
delete_task

Move a task to trash (recoverable).
  • id (string): Task ID

​
get_config

Retrieve all available values for assignees, dartboards, statuses, tags, custom properties, etc.

​
get_dartboard

Retrieve a dartboard by ID.
  • id (string): Dartboard ID

​
get_doc

Retrieve a doc by ID.
  • id (string): Doc ID

​
get_folder

Retrieve a folder by ID.
  • id (string): Folder ID

​
get_task

Retrieve a task by ID.
  • id (string): Task ID

​
get_view

Retrieve a view by ID.
  • id (string): View ID

​
list_docs

List docs with filters: folder, title, text, etc.

​
list_task_comments

List comments with filters: author, task, text, date, etc.

​
list_tasks

List tasks with filters: assignee, status, dartboard, priority, dates, tags, etc.

​
update_doc

Update a doc’s title, content, or folder.
  • id (string): Doc ID

​
update_task

Update all properties of a task.
  • id (string): Task ID

​
Meshagent Resources

For more details about Dart MCP Server see: Run securely: Why use Docker for MCP?