Skipprd discover

Connect to the data source, sample records, and infer the pipeline schema. Unlike sync, discover never writes to the output destination -- it only discovers schemas and persists metadata.

Usage

skipprd discover --pipeline <name> [--output <mode>] [--log [LEVEL]]
skipprd --config skippr.yml discover --pipeline <name> [--output <mode>] [--log [LEVEL]]

Flags

Flag	Required	Description
--pipeline, -p	No	Pipeline name. Falls back to PIPELINE_NAME env var.
--output	No	Output mode: progress (default, interactive spinner), json (structured JSON lines to stdout), or text (plain text summaries).
--log	No	Enable logging. Optional level: debug, info, warn, error. Defaults to info when flag is present.

What it does

1. Loads or creates pipeline metadata (from S3 or local disk, depending on SKIPPRD_EL_STORAGE_MODE)
2. Initializes the offset database
3. Connects to the data source configured in skippr.yml
4. Samples records and infers the complete nested schema via type inference
5. Persists the updated pipeline metadata
6. Exits

Discover does not initialize or sync to any output plugin. It is purely a schema inference operation.

Example

skipprd --config skippr.yml discover --pipeline events --log

Structured output for programmatic use

skipprd discover --pipeline el_mssql --output json