Catalogue

🤖 AI Drafted (Not reviewed)

Run the whole Catalogue pipeline: the six stage presets in this folder, in order, as one chain. Import → Artifacts registers per-page import receipts and transcription artifacts from the already-imported page content; Extract Entities and Extract SVO → Claims read those artifacts and persist entity and claim rows ($small model); Merge / Dedup reapplies curation rules; KG Persist / Finalize recomputes corroboration, embeddings and the graph snapshot; Catalogue Description writes the narrative archival description ($small model). Every stage is also runnable standalone from this folder, and each is safe to re-run. Pages must already carry text — run a Transcribe workflow first for scans without content.

Folder	/Catalogue
Steps	7
Tags	preset, catalogue, chain, pipeline, sub-workflow, small

Steps, in run order

1. Files

Tool: Files — Pass through input files from workflow context

2. Import → Artifacts

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}]
output_contract	[{“data_type”: “json”, “description”: “Stage summary; also the ordering edge into stage 2.”, “id”: “summary”, “required”: true}]
output_mapping	{“summary”: “$.nodes.import-artifacts.summary”}
workflow_ref	Import → Artifacts

3. Extract Entities

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}, {“data_type”: “any”, “description”: “Ordering edge: stage 1 must finish before entities are extracted.”, “id”: “barrier”, “required”: false}]
output_contract	[{“data_type”: “json”, “description”: “Stage summary; also the ordering edge into stage 3.”, “id”: “summary”, “required”: true}]
output_mapping	{“summary”: “$.nodes.extract-entities.summary”}
workflow_ref	Extract Entities

4. Extract SVO → Claims

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}, {“data_type”: “any”, “description”: “Ordering edge: stage 2’s entity rows must exist before claims are extracted.”, “id”: “barrier”, “required”: false}]
output_contract	[{“data_type”: “json”, “description”: “Stage summary; also the ordering edge into stage 4.”, “id”: “summary”, “required”: true}]
output_mapping	{“summary”: “$.nodes.extract-svo.summary”}
workflow_ref	Extract SVO → Claims

5. Merge / Dedup

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}, {“data_type”: “any”, “description”: “Ordering edge: stages 2-3 must have written their rows before curation reapplies.”, “id”: “barrier”, “required”: false}]
output_contract	[{“data_type”: “json”, “description”: “Stage summary; also the ordering edge into stage 5.”, “id”: “summary”, “required”: true}]
output_mapping	{“summary”: “$.nodes.merge-dedup.summary”}
workflow_ref	Merge / Dedup

6. KG Persist / Finalize

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}, {“data_type”: “any”, “description”: “Ordering edge: merged/deduped rows must be final before corroboration and the graph snapshot.”, “id”: “barrier”, “required”: false}]
output_contract	[{“data_type”: “json”, “description”: “Stage summary; also the ordering edge into stage 6.”, “id”: “summary”, “required”: true}]
output_mapping	{“summary”: “$.nodes.kg-persist-finalize.summary”}
workflow_ref	KG Persist / Finalize

7. Catalogue Description

Tool: Sub-Workflow — Run a child workflow behind declared input/output contracts.

Settings this step uses:

Option	Value
input_contract	[{“data_type”: “any”, “description”: “Selected document metadata (scope is re-resolved from the selection inside the stage).”, “id”: “documents”, “required”: false}, {“data_type”: “any”, “description”: “Ordering edge: the narrative reads the KG rows stages 1-5 produced.”, “id”: “barrier”, “required”: false}]
output_contract	[{“data_type”: “text”, “description”: “The narrative catalogue entry written onto the selected container.”, “id”: “text”, “required”: true}]
output_mapping	{“text”: “$.nodes.catalogue.text”}
workflow_ref	Catalogue Description