K3 Pipeline Frontend

1) Run pipeline

Config template path

Output path

Pipeline dataset identifier (shared across staged sources)

Existing profile files

Data file keys (auto-detected raw input files) i Read-only raw files auto-detected from the latest pipeline run.

Run job id: not run yet Data file keys: none i These keys are auto-detected from the latest pipeline run. manifest.json is intentionally filtered out because it is metadata, not raw input data. page_0001.json and page_0002.json are raw staged pages. trials_min.jsonl is extracted normalized records used for downstream profiling/reconciliation.

Registered services execution

Load services to view per-service execution status.

2) Visualise RDF

RDF files from current run

Select one or more files (Cmd/Ctrl + click for multi-select), then load/visualise together.

Selected RDF Turtle URL

Visualisation service

1b) Pipeline organizer

Load registered services to organize execution order and handoffs.

Pipeline service order (first -> last)

Add service to order

Service handoff lambdas (output -> next input)

Input/output per service (latest run)

Run pipeline to inspect chained input/output per service.

K3 Pipeline Frontend

1) Run pipeline

2) Visualise RDF

1a) Service I/O Inspector

1b) Pipeline organizer

SSSOM Template Editor

Token session required

Service run details

K3 Competency Questions