graph TD %%{init: {'theme': 'mc','layout': 'elk'}}%% ParseData-ptp7g[<div><img alt="logo" src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Obtener Datos] style ParseData-ptp7g stroke:#a170ff Prompt-7qn0s[<div><img alt="logo" src="/_astro/square-terminal.BMOXc-nZ.svg" style="height: 20px !important;width: 20px !important"/></div>Consulta] style Prompt-7qn0s stroke:#a170ff OpenAIModel-kx9h5[<div><img alt="logo" src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI] style OpenAIModel-kx9h5 stroke:#a170ff ChatOutput-ugeve[<div><img alt="logo" src="/_astro/messages-square.BaSDmT6g.svg" style="height: 20px !important;width: 20px !important"/></div>Chat Output] style ChatOutput-ugeve stroke:#a170ff GmailMesageComponent-riqjp[<div><img alt="logo" src="/_astro/gmail.CLxyIGOu.svg" style="height: 20px !important;width: 20px !important"/></div>Obtener Mensajes] style GmailMesageComponent-riqjp stroke:#a170ff GmailMesageComponent-1gc2u[<div><img alt="logo" src="/_astro/gmail.CLxyIGOu.svg" style="height: 20px !important;width: 20px !important"/></div>Marcar como leído] style GmailMesageComponent-1gc2u stroke:#a170ff TextInput-1cb9o[<div><img alt="logo" src="/_astro/type.Dy26vmDy.svg" style="height: 20px !important;width: 20px !important"/></div>Tema a comprobar] style TextInput-1cb9o stroke:#a170ff AdvancedAgent-csduv[Agent2] style AdvancedAgent-csduv stroke:#a170ff GmailMesageComponent-riqjp -.- ParseData-ptp7g linkStyle 0 stroke:#a170ff ParseData-ptp7g -.- Prompt-7qn0s linkStyle 1 stroke:#a170ff TextInput-1cb9o -.- Prompt-7qn0s linkStyle 2 stroke:#a170ff Prompt-7qn0s -.- AdvancedAgent-csduv linkStyle 3 stroke:#a170ff AdvancedAgent-csduv -.- ChatOutput-ugeve linkStyle 4 stroke:#a170ff OpenAIModel-kx9h5 -.- AdvancedAgent-csduv linkStyle 5 stroke:#a170ff GmailMesageComponent-1gc2u -.- AdvancedAgent-csduv linkStyle 6 stroke:#a170ff

🧩 Unread Message Processor

Overview
The Unread Message Processor streamlines the handling of unread emails by automatically retrieving messages that match a user‑defined query, parsing their content into a readable format, and leveraging an AI agent to decide which messages should be marked as read. The workflow uses a GPT‑powered agent that can interact with Gmail via a dedicated tool, finally presenting the agent’s decision in the Playground chat interface.

⚙️ Main Features

Retrieves Gmail messages that satisfy a user‑defined query.
Converts raw email data into structured plain text using a custom template.
Builds a prompt that contains the email list and the user‑specified topic.
Employs a GPT‑4 language model to determine which messages to mark as read.
Calls a Gmail tool to mark the selected messages as read.
Displays the AI’s final response in a chat‑style output.

🔄 Workflow Steps

Component Name	Role in the Workflow	Key Inputs	Key Outputs
Text Input	Captures the topic to filter emails (e.g., “LinkedIn”).	Text (topic)	Text message containing the topic
Gmail Message Component	Fetches Gmail messages that match the query and returns raw email data.	Query string (“LinkedIn”), Gmail credentials, number of emails	Raw email data (Data)
Parse Data	Formats each email into a plain‑text representation based on a template.	Raw email data	Structured text of emails
Prompt	Constructs the prompt that will be fed to the agent, embedding the list of emails and the topic.	Structured email text, topic	Prompt message
OpenAI Model	Provides the language model that powers the agent’s reasoning.	Model selection (e.g., GPT‑4.1), API credentials	LanguageModel instance
Advanced Agent	Executes the main decision logic, using the prompt, the language model, and the Gmail tool.	Prompt, LanguageModel, Gmail tool	Response message
Gmail Message Component	Marks the chosen emails as read via the tool exposed by the agent.	List of email IDs to mark, Gmail credentials	Confirmation data
Chat Output	Displays the agent’s response in the Playground chat window.	Response message	Chat message shown to the user

🧠 Notes

The workflow requires valid Gmail API credentials for both the retrieval and marking tools.
An OpenAI API key is mandatory for the language model component.
The prompt template expects a list of emails; if no emails match the query, the agent will handle the empty case gracefully.
The agent’s decision logic can be tuned via the system prompt and the structured output schema, if needed.
The Gmail marking tool operates only on the emails returned in the retrieval step; it does not affect other messages.
The chat output component displays the agent’s final decision in a conversational format, suitable for immediate review.