graph TD %%{init: {'theme': 'mc','layout': 'elk'}}%% TextInput-7a156[<div><img src="/_astro/type.Dy26vmDy.svg" style="height: 20px !important;width: 20px !important"/></div>Instructions] style TextInput-7a156 stroke:#a170ff ChatOutput-ol8bm[<div><img src="/_astro/messages-square.BaSDmT6g.svg" style="height: 20px !important;width: 20px !important"/></div>Chat Output] style ChatOutput-ol8bm stroke:#a170ff OpenAIModel-d3ck8[<div><img src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI] style OpenAIModel-d3ck8 stroke:#a170ff TextInput-cv9yq[<div><img src="/_astro/type.Dy26vmDy.svg" style="height: 20px !important;width: 20px !important"/></div>ImageURL] style TextInput-cv9yq stroke:#a170ff DescribeImage-kmqoa[<div><img src="/_astro/image.segxnpQg.svg" style="height: 20px !important;width: 20px !important"/></div>Describe Image] style DescribeImage-kmqoa stroke:#a170ff ParseData-7m4sd[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>DescriptionExtraction] style ParseData-7m4sd stroke:#a170ff OpenAIModel-d3ck8 -.- DescribeImage-kmqoa linkStyle 0 stroke:#a170ff TextInput-7a156 -.- DescribeImage-kmqoa linkStyle 1 stroke:#a170ff DescribeImage-kmqoa -.- ParseData-7m4sd linkStyle 2 stroke:#a170ff ParseData-7m4sd -.- ChatOutput-ol8bm linkStyle 3 stroke:#a170ff TextInput-cv9yq -.- DescribeImage-kmqoa linkStyle 4 stroke:#a170ff

Ticket to JSON

🧩 Overview

The Ticket to JSON workflow automates the extraction of structured data from digital images of fast‑food restaurant post‑sale tickets. By combining a prompt‑based text instruction, an image URL, an OpenAI language model, and a series of transformation components, the system produces a JSON representation of the ticket contents and presents the result in a chat interface.

⚙️ Main Features

Accepts a predefined extraction prompt and an image URL as input.
Utilises an OpenAI language model to generate a natural‑language description of the image.
Parses the description into plain text using a user‑defined template.
Outputs the final JSON payload in a conversational chat.
Supports easy substitution of the underlying language model.

🔄 Workflow Steps

Component Name	Role in the Workflow	Key Inputs	Key Outputs
Instructions	Provides the extraction prompt to guide the model.	None	Prompt Text
Image URL	Supplies the ticket image to be processed.	None	Image Data
OpenAI	Supplies a language model instance for image analysis.	Prompt Text, Image Data	Model (LanguageModel)
Describe Image	Generates a textual description of the image using the LLM.	Image Data, Prompt Text, LLM	Describe Data (Structured Description)
Description Extraction	Converts the description data into plain text using a template.	Describe Data, Template	Text (Parsed Output)
Chat Output	Displays the parsed text to the user in chat.	Text	Chat Message (JSON Output)

🧠 Notes

The workflow requires a publicly accessible image URL; it does not handle local file uploads.
The extraction prompt is hard‑coded in the Instructions component but can be modified to target other document types.
The OpenAI component can be configured to use any available OpenAI model (e.g., gpt‑4o, gpt‑5).
JSON mode is disabled in the OpenAI component; the final JSON structure is constructed in the Description Extraction step.
The system assumes internet connectivity for API calls and image retrieval.
The Chat Output component presents the result as a chat message; it can be redirected to other interfaces if required.
No internal IDs or configuration details are exposed in the documentation, ensuring clarity and privacy.