How princeg0210/autonomous-catalog-enrichment works
This repository implements a product-data enrichment pipeline with a small dashboard/API front end. The principal workflow accepts a SKU, dispatches asynchronous enrichment, builds an isolated product context, discovers and classifies source data, then validates and persists enriched results or review exceptions. The README documents additional source tiers and syndication capabilities beyond the sampled implementation. Edges use excerpts, SOURCE INDEX references, or README statements; unsampled subsystem details and integrations are not expanded beyond that evidence.
Main entry points: tasks.py Async Enrichment · discovery.py Source Discovery · main.py API Gateway · models.py Database Operations
Components 18
Ingestion & UI
- API Gateway
main.py - Web Dashboard
static/index.html - Redis Cache
Enrichment Pipeline
- Async Enrichment
tasks.py - Product Context
context.py - Source Discovery
discovery.py - Catalog Dataset
dataset.csv - Product Sources
- Attribute Classification
attribute_classifier.py - Unit Normalization
normalizer.py - Attribute Deduplication
deduplicator.py - Taxonomy Classification
taxonomy.py
Quality & Syndication
- Channel Copywriting
copywriter.py - Quality Gate
quality_gate.py
Persistence & Review
- Database Operations
models.py - Product Database
- HITL Queue
Other components
- Catalog Operator
Connections 20
- Catalog Operator → API Gateway: submits SKU
README.md - API Gateway → Redis Cache: checks cache
main.py - API Gateway → Async Enrichment: dispatches job
main.py - API Gateway → Web Dashboard: serves dashboard
main.py - Web Dashboard → API Gateway: requests data
- Async Enrichment → Redis Cache: checks cache
tasks.py - Async Enrichment → Product Context: creates context
tasks.py - Async Enrichment → Source Discovery: runs discovery
tasks.py - Source Discovery → Catalog Dataset: reads records
discovery.py - Source Discovery → Product Sources: fetches product page
discovery.py - Source Discovery → Attribute Classification: classifies attributes
discovery.py - Source Discovery → Unit Normalization: normalizes values
discovery.py - Source Discovery → Attribute Deduplication: deduplicates attributes
discovery.py - Async Enrichment → Taxonomy Classification: classifies product
tasks.py - Async Enrichment → Channel Copywriting: generates descriptions
tasks.py - Async Enrichment → Quality Gate: evaluates quality
tasks.py - Async Enrichment → Database Operations: persists results
tasks.py - Database Operations → Product Database: reads and writes
models.py - Async Enrichment → HITL Queue: records exceptions
tasks.py - API Gateway → Database Operations: lists review items
main.py