Skip to content
Floe

Plugin Configuration and CLI

This document describes plugin CLI commands and how to create custom plugins.

Plugin CLI Commands

Section titled “Plugin CLI Commands”
Terminal window
# List installed plugins
floe plugins list


# Illustrative output:
Installed plugins:
  orchestrators:
    - dagster (1.0.0) [reference implementation]
    - airflow (1.0.0)
  computes:
    - duckdb (1.0.0) [reference implementation]
    - snowflake (1.0.0)
    - spark (1.0.0)
  catalogs:
    - polaris (1.0.0) [reference implementation]
    - glue (1.0.0)
  dbt:
    - local (1.0.0) [alpha-supported local runtime]
    - fusion (1.0.0)
  semantic_layers:
    - cube (1.0.0) [reference implementation]
    - none (1.0.0)
  ingestion:
    - dlt (1.0.0) [implementation primitive]
    - airbyte (1.0.0)


# List available (installable) plugins
floe plugins available

Creating a Custom Plugin

Section titled “Creating a Custom Plugin”

1. Create Package Structure

Section titled “1. Create Package Structure”
Terminal window
mkdir floe-compute-trino
cd floe-compute-trino

2. Implement Interface

Section titled “2. Implement Interface”
src/floe_compute_trino/plugin.py
from floe_core.interfaces.compute import ComputePlugin, ComputeConfig


class TrinoComputePlugin(ComputePlugin):
    name = "trino"
    version = "1.0.0"
    is_self_hosted = True


    metadata = PluginMetadata(
        name="trino",
        version="1.0.0",
        floe_api_version="1.0",
        description="Trino compute plugin for floe",
        author="Your Name",
    )


    def generate_dbt_profile(self, config: ComputeConfig) -> dict:
        return {
            "type": "trino",
            "method": "none",
            "host": config.properties.get("host", "trino.default.svc.cluster.local"),
            "port": config.properties.get("port", 8080),
            "catalog": config.properties.get("catalog", "iceberg"),
            "schema": config.properties.get("schema", "default"),
        }


    def get_required_dbt_packages(self) -> list[str]:
        return ["dbt-trino>=1.7.0"]


    # ... implement other methods

3. Register Entry Point

Section titled “3. Register Entry Point”
pyproject.toml
[project.entry-points."floe.computes"]
trino = "floe_compute_trino:TrinoComputePlugin"

4. Add Helm Chart (if needed)

Section titled “4. Add Helm Chart (if needed)”
chart/Chart.yaml
apiVersion: v2
name: floe-compute-trino
version: 1.0.0
description: Trino compute for floe


# chart/templates/deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: trino-coordinator
# ...

5. Install and Use

Section titled “5. Install and Use”
Terminal window
uv add floe-compute-trino


# manifest.yaml
plugins:
  compute:
    type: trino
    config:
      host: trino.example.com
Section titled “Related Documents”