Dlt
CertifiedTasks that run dlt (data load tool) pipelines and CLI commands to move data into analytics destinations.
Use these tasks to execute dlt workloads either from inline Python (Run) or through non-interactive CLI commands (CLI). Provide a Python script or list of commands, set beforeCommands to install dependencies, and adjust the default ghcr.io/kestra-io/dlt image when you need custom environments; the CLI task also auto-adds --non-interactive and installs requirements.txt after dlt init.
Dlt
Tasks that run dlt (data load tool) pipelines and CLI commands to move data into analytics destinations.
Use these tasks to execute dlt workloads either from inline Python (Run) or through non-interactive CLI commands (CLI). Provide a Python script or list of commands, set beforeCommands to install dependencies, and adjust the default ghcr.io/kestra-io/dlt image when you need custom environments; the CLI task also auto-adds --non-interactive and installs requirements.txt after dlt init.
tasks
How to use the dlt plugin
Run dlt (data load tool) pipelines from Kestra flows to extract and load data across sources and destinations.
Authentication
dlt pipelines authenticate to sources and destinations via environment variables. Pass credentials through the env map using dlt's nested variable convention (e.g. SOURCES__<SOURCE>__<SECTION>__CREDENTIALS__<KEY>). Store secrets in secrets and apply connection properties globally with plugin defaults.
Tasks
CLI runs dlt CLI commands in a container (default image ghcr.io/kestra-io/dlt) — set commands (required, list of shell commands such as dlt init, dlt pipeline, or dlt run). Use beforeCommands for setup steps, env for credentials, inputFiles to stage local files, and outputFiles to retrieve results.
Run executes an inline Python dlt script — set script (required, a Python string that defines and runs the pipeline). Use beforeCommands, env, inputFiles, and outputFiles the same way as CLI. The default container image is also ghcr.io/kestra-io/dlt.
