Infinitely scalable orchestration platform.
## Kestra MCP Server: Orchestration & Data Pipeline The **Kestra MCP Server** integrates Kestra's unified orchestration platform into Google Antigravity, enabling developers to build, schedule, and monitor complex data pipelines and workflows through AI-assisted development. ### Why Kestra MCP? - **Declarative Workflows**: Define pipelines as code using YAML with 400+ plugins - **Event-Driven Architecture**: Trigger workflows based on events, schedules, or API calls - **Real-Time Monitoring**: Track execution status, logs, and metrics in real-time - **Version Control**: Built-in Git integration for workflow versioning and collaboration - **Multi-Cloud Support**: Run workflows across AWS, GCP, Azure, and on-premises ### Key Features #### 1. Workflow Definition and Execution ```yaml # Define workflows via MCP id: data_pipeline namespace: analytics tasks: - id: extract type: io.kestra.plugin.jdbc.Query sql: SELECT * FROM source_table WHERE updated_at > :lastRun - id: transform type: io.kestra.plugin.scripts.python.Script script: | import pandas as pd df = pd.DataFrame(inputs) df_cleaned = transform_data(df) return df_cleaned.to_dict() - id: load type: io.kestra.plugin.gcp.bigquery.Load table: analytics.processed_data triggers: - id: daily_schedule type: io.kestra.core.models.triggers.types.Schedule cron: "0 6 * * *" ``` #### 2. Pipeline Monitoring ```python # Monitor workflow executions executions = await kestra.listExecutions( namespace="analytics", flow_id="data_pipeline", state="RUNNING" ) # Get execution logs for execution in executions: logs = await kestra.getExecutionLogs(execution.id) metrics = await kestra.getExecutionMetrics(execution.id) ``` ### Configuration ```json { "mcpServers": { "kestra": { "command": "npx", "args": ["-y", "@anthropic/mcp-kestra"], "env": { "KESTRA_API_URL": "https://your-kestra-instance.com", "KESTRA_API_TOKEN": "your-api-token" } } } } ``` ### Use Cases **ETL Pipelines**: Build robust extract-transform-load pipelines with error handling, retries, and notifications. **ML Workflow Orchestration**: Orchestrate machine learning pipelines from data preparation through model deployment. **Event-Driven Automation**: Create reactive workflows that respond to business events like new orders or user signups. The Kestra MCP Server brings enterprise-grade orchestration capabilities to your AI-assisted development workflow.
{
"mcpServers": {
"kestra": {}
}
}