Chat Gpt - PlantonCloud

1. Adding Credentials to the Platform

Navigate to the Connections Page
- On the Connections page, you will see a set of connection cards for different providers.
Locate the OpenAI (ChatGPT) Card
- Look for the card labeled OpenAI or ChatGPT.
- This card represents the connection needed to enable ChatGPT services within the platform.
Click “Connect”
- Clicking the Connect button on the card will open a form that requests specific details needed to establish the ChatGPT integration.
Fill Out the ChatGPT Form
- Provide the required information as outlined below.
- Once you have completed all fields, click Submit to finalize the connection.

2. Form Details of the Credential

When you open the Create ChatGPT form (as shown in the provided screenshot), there are two main sections: Metadata and OpenAI Account. Each section contains fields that must be filled in to successfully integrate ChatGPT.

2.1 Metadata

Name
- Description: A user-friendly identifier or nickname for this ChatGPT credential.
- Purpose: This name helps you recognize and distinguish this credential from others you may have.
- Example: “Marketing-ChatGPT” or “Internal-QA-ChatGPT”

2.2 OpenAI Account

Within the OpenAI Account section, you will see the following fields:

API Key
- Description: The secret key used to authenticate your API calls to OpenAI.
- Purpose: Allows the platform to securely communicate with the OpenAI API on your behalf.
- How to Obtain:
  1. Log in to your OpenAI Dashboard.
  2. Navigate to View API keys under your account settings.
  3. Click Create new secret key and copy the generated key.
  4. Paste the key into the API Key field in the platform’s form.
- Important: Treat your API Key like a password—never share it publicly or store it in unsecured locations.
API URL
- Description: The endpoint URL to which all requests for ChatGPT will be sent.
- Purpose: Ensures your API calls go to the correct service endpoint (e.g., OpenAI’s default API endpoint, or a custom endpoint if you are using Azure OpenAI).
- Common Example: https://api.openai.com/v1
- How to Obtain:
  - If you are using standard OpenAI services, use https://api.openai.com/v1.
  - If you have an Azure OpenAI deployment, retrieve the endpoint from your Azure portal.
Model ID
- Description: The identifier of the specific ChatGPT model you want to use.
- Purpose: Informs the platform which version or variant of OpenAI’s language model to call.
- Common Examples:
  - gpt-3.5-turbo
  - gpt-4 (if you have access)
- How to Obtain:
  - From your OpenAI dashboard, review the list of models you have access to.
  - If using Azure OpenAI, confirm the exact deployment name or model ID in your Azure portal.

Additional Notes

Clarity & Consistency
- Keep your naming conventions and settings uniform across all environments to avoid confusion.
Security Best Practices
- Only share or store API credentials in secure, encrypted vaults or within your organization’s approved secret management tool.
User-Friendly Terminology
- When documenting for team members, avoid overly technical jargon. Use clear, concise language to help everyone understand how to configure ChatGPT.
Provider-Specific Guidelines
- If using providers other than OpenAI (such as Azure), be sure to follow their portal instructions for generating API keys, setting endpoints, and specifying model names.
Version Control
- Because OpenAI’s models and endpoints can change over time, periodically review your model IDs and endpoints to keep them up to date.

That’s it! Once you’ve completed and submitted this form, your ChatGPT integration should be ready to use in your workflows. Make sure to test your credentials by sending a sample request to confirm everything is working properly. If you encounter any errors, double-check your API key, endpoint, and model ID against the details in your OpenAI or Azure portal.