Durable workflow execution platform.
## Temporal MCP Server: Durable Workflow Orchestration The **Temporal MCP Server** integrates the durable execution platform into Google Antigravity, enabling developers to build and manage reliable, long-running workflows that survive failures. ### Why Temporal MCP? - **Durable execution** - Workflows survive process failures - **Workflow control** - Start, signal, and query workflows - **History access** - Full execution history for debugging - **Retry handling** - Automatic retry with backoff - **Visibility** - Search and filter workflows ### Key Features #### 1. Workflow Execution ```python # Start workflow execution = await mcp.call("temporal", "start_workflow", { "workflow_type": "ProcessOrder", "workflow_id": "order-12345", "task_queue": "orders", "input": {"order_id": "12345", "items": ["item1", "item2"]} }) print(f"Run ID: {execution[\"run_id\"]}") # Get workflow result result = await mcp.call("temporal", "get_workflow_result", { "workflow_id": "order-12345" }) ``` #### 2. Workflow Control ```python # Signal workflow await mcp.call("temporal", "signal_workflow", { "workflow_id": "order-12345", "signal_name": "update_shipping", "input": {"address": "123 Main St"} }) # Query workflow state = await mcp.call("temporal", "query_workflow", { "workflow_id": "order-12345", "query_type": "get_status" }) # Cancel workflow await mcp.call("temporal", "cancel_workflow", { "workflow_id": "order-12345" }) ``` #### 3. Workflow History ```python # Get execution history history = await mcp.call("temporal", "get_history", { "workflow_id": "order-12345" }) for event in history["events"]: print(f"{event[\"event_type\"]}: {event[\"timestamp\"]}") ``` #### 4. Workflow Search ```python # List workflows by type and status workflows = await mcp.call("temporal", "list_workflows", { "workflow_type": "ProcessOrder", "status": "Running" }) for wf in workflows: print(f"{wf[\"workflow_id\"]}: {wf[\"status\"]}") ``` ### Configuration ```json { "mcpServers": { "temporal": { "command": "npx", "args": ["-y", "@anthropic/mcp-temporal"], "env": { "TEMPORAL_ADDRESS": "localhost:7233", "TEMPORAL_NAMESPACE": "default" } } } } ``` ### Use Cases **Order Processing**: Reliable order fulfillment workflows. **Data Pipelines**: Long-running ETL processes. **Sagas**: Distributed transaction coordination. **Scheduled Jobs**: Durable scheduled task execution. The Temporal MCP Server enables durable workflow orchestration.
{
"mcpServers": {
"temporal": {}
}
}