mirror of
https://github.com/R0m1k3/Priceflow.git
synced 2026-10-11 17:29:14 +02:00
1.4 KiB
1.4 KiB
Walkthrough: Vision-First Price Extraction
In response to issues where the AI was being misled by hidden text (like unit prices or old prices in HTML), we have implemented a Vision-Priority Strategy.
Changes Implemented
1. Updated AI System Prompt (app/ai_schema.py)
We completely rewrote the EXTRACTION_PROMPT_TEMPLATE to enforce the following rules:
- Source of Truth = Image: explicit instruction that the screenshot takes precedence over any text.
- Conflict Resolution: "IF IMAGE AND TEXT CONFLICT, TRUST THE IMAGE."
- Visual Focus Rules:
- Look for the largest/boldest price.
- Ignore small, styling-less text (often unit prices).
- Ignore crossed-out text.
Verification
We verified the new prompt generation using verify_vision_priority.py.
Generated Prompt Preview:
You are a Vision-First Price Extraction Agent.
Your Goal: Extract the main product price exactly as a human sees it on the screen.
**SOURCE OF TRUTH = IMAGE**
- The image provided is the **Absolute Truth**.
- The text provided below is scraped HTML content which may contain hidden/old prices.
- **IF IMAGE AND TEXT CONFLICT, TRUST THE IMAGE.**
How to Test
- Go to "Suivis Prix".
- Force refresh an item that was previously incorrect (e.g., B&M item showing unit price).
- The AI should now ignore the "hidden" unit price text and read the main price tag from the image.