Build a Merchant Onboarding Evidence Workspace with Fictional Checklists

Extract identifiers from unstructured notes against fictional onboarding checklist rules. Output normalized draft records for human review before conversion to Beckn catalog schemas.

Seller network onboarding specialist

Ingest informal merchant product descriptions and mock checklist configuration rules.

Extract raw text fields like item titles, prices, and alleged license numbers.

Perform format-only validation checks against the provided fictional checklist requirements.

Generate a normalized draft record that mandates human review before a separate versioned Beckn adapter handles schema conversion.

Split-screen review workbench showing source text on the left and extracted draft records on the right for human approval.

Construct synthetic item descriptions and fictional mock checklist JSON rule files for extraction targets. Starter examples below are synthetic. They are not downloaded records.

Beckn protocol specifications Utilizes Beckn protocol core catalog discovery specifications as the target format for approved adapter exports.

pydantic-ai, fastapi, pandas. Pydantic AI provides structured extraction capabilities for parsing raw text into normalized draft schema objects based on custom checklist rules. Local or hosted LLM with strong extraction and structured object generation capabilities.

Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

Extracts specified product details into a draft schema without asserting legal compliance.

Flags records lacking fictional checklist items for human review instead of triggering automatic rejection.

Never outputs a COMPLIANT status or automatic onboarding approval for any generated record.

Requires explicit human-in-the-loop signoff on the generated draft outputs.

```json [ { "input": { "raw_text": "Mustard Oil 1L. MRP 240. Lic No: 12345678901234.", "checklist_keys": [ "name", "price", "license_code" ] }, "expected_output": { "draft_record": { "name": "Mustard Oil 1L", "price": "240", "license_code_extracted": "12345678901234", "format_check": "passed" }, "requires_human_review": true, "note": "Simplified synthetic app record, not a complete official schema file." } }, { "input": { "raw_text": "Pickle Jar. Very tasty mango.", "checklist_keys": [ "name", "price", "license_code" ] }, "expected_output": { "draft_record": { "name": "Pickle Jar", "price": null, "license_code_extracted": null, "format_check": "missing_fields" }, "requires_human_review": true, "note": "Simplified synthetic app record, not a complete official schema file." } } ] ```

Evaluates synthetic item catalogs; does not connect to live commercial marketplaces or collect proprietary banking data. Use mock external writes. Require human review before any real deployment.