Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@task-manager Create a new step named 'data validation' for execution exec-789"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Nova Agent progress reporting MCP server.
./build-docker.shmacOS/Windows:
claude mcp add task-manager -s user \
--env "TASK_MANAGER_HOST=host.docker.internal" \
--env "TASK_MANAGER_PORT=8080" \
--env "TASK_MANAGER_TIMEOUT=30" \
--env "USE_MOCK_CLIENT=false" \
-- docker run -i --rm \
-e TASK_MANAGER_HOST \
-e TASK_MANAGER_PORT \
-e TASK_MANAGER_TIMEOUT \
-e USE_MOCK_CLIENT \
task-manager-mcp:latestLinux:
claude mcp add task-manager -s user \
--env "TASK_MANAGER_HOST=localhost" \
--env "TASK_MANAGER_PORT=8080" \
--env "TASK_MANAGER_TIMEOUT=30" \
--env "USE_MOCK_CLIENT=false" \
-- docker run -i --rm --network=host \
-e TASK_MANAGER_HOST \
-e TASK_MANAGER_PORT \
-e TASK_MANAGER_TIMEOUT \
-e USE_MOCK_CLIENT \
task-manager-mcp:latestMock Mode (for testing):
claude mcp add task-manager-mock -s user \
--env "USE_MOCK_CLIENT=true" \
-- docker run -i --rm \
-e USE_MOCK_CLIENT \
task-manager-mcp:latestpip install -r requirements.txtEdit Claude Desktop's MCP configuration file (refer to mcp-config-example.json):
{
"mcpServers": {
"task-manager": {
"command": "python3",
"args": ["/path/to/your/task_manager_mcp.py"],
"env": {
"TASK_MANAGER_HOST": "localhost",
"TASK_MANAGER_PORT": "8080",
"TASK_MANAGER_TIMEOUT": "30",
"USE_MOCK_CLIENT": "false"
}
}
}
}# Remove MCP server
claude mcp remove task-manager
# Or remove mock version
claude mcp remove task-manager-mockRelated MCP server: Cairn
Tool | Purpose |
| Update execution's session_id |
| Create a step and return step_id |
| Update step status/message |
| Health check |
# 1. Register session
update_execution_session(execution_id="exec-123", session_id="nova-001")
# 2. Create step
result = create_step(execution_id="exec-123", step_name="analyzing")
step_id = result["step_id"]
# 3. Complete step
update_step(execution_id="exec-123", step_id=step_id, status="completed")running - In progress
completed - Completed
failed - Failed
skipped - Skipped
Variable | Description | Default |
| Task Manager service address |
|
| Task Manager service port |
|
| Request timeout (seconds) |
|
| Whether to use Mock client |
|
macOS/Windows: Use host.docker.internal to access host services
Linux: Use --network=host and localhost to access host services
pip install -r requirements.txt
python task_manager_mcp.pymake test
# or
python tests/simple_test.py
python tests/test_generated_client.pyWhen backend API is unavailable, MCP returns clear error messages:
404 Error (API not implemented):
{
"success": false,
"error": "API endpoint not found: PATCH /api/executions/{id}/steps/{step_id}. The backend service may not have implemented this API yet.",
"status_code": 404,
"hint": "Please check if the Task Manager backend service has this endpoint implemented."
}Connection Failed:
{
"success": false,
"error": "Cannot connect to Task Manager service at http://localhost:8080",
"hint": "Please verify that the Task Manager service is running and the host/port are correct."
}This server cannot be deployed
MCP Server for an Agent Task Marketplace
Hosted MCP server for task-first delegation to remote workstations and workers.
MCP server for mandates, delegation, policy-gated execution, credential grants, and audit.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
