Usage Guide

Disclaimer

• This service uses AI models in the backend and responses may not always be accurate. Please verify important information independently and use at your own discretion.
• The token count displayed after each response represents the total tokens consumed (input + output combined), not just the output tokens.
• Chat data is stored in your current browser session only. All conversations will be lost when you close the tab or sign out.
• If you need to keep a response, use the download button ( ⇩ ) available on each AI response to export it as TXT, PDF, Word, or Excel before ending your session.
• The like and dislike options after chat responses are not recorded at this time.

Upcoming Enhancements

The following features are actively being developed by Ravi Kumar and will be available in future releases.

• Docs and Excel export support is under development and will be available in an upcoming release.
• Media output (images, visual content) support is planned and currently being implemented.
• If an API key expires or is not responding, please try selecting a different API key from the dropdown.
• Some API keys may have a different prefix depending on the provider.
• A valid API key is required to use the chat functionality.

Working Functionality

Login

• Open the application URL in your browser.
• Sign in using your credentials (Microsoft Entra ID or demo login depending on configuration).
• After successful login, you will be redirected to the main chat screen.

Main Chat Screen

• Select an AI model from the Model dropdown (top-left).
• Select an API key / account from the API Key / Account dropdown.
• Type your message in the text box at the bottom and click Send.
• The AI response will appear below your message along with the model name, account name, and total tokens consumed.
• You can use the Like or Dislike buttons on any response (note: these are not recorded at this time).
• You can click Reveal Key to view the full API key and endpoint for the selected account.

Chat Sessions

• Click + Start New Chat to begin a fresh conversation.
• Previous chats appear under Recent Chats in the sidebar. Click any to resume.
• Each browser tab maintains its own independent chat history.

File Attachments

• Click the Attach button to upload files, or drag and drop files into the chat area.
• Supported file types: PDF, DOC, DOCX, PPT, PPTX, XLS, XLSX, CSV, TXT, MD, JPEG, JPG, PNG, MP4.
• Maximum 5 files per message, each up to 2 MB.
• Attached files are sent to the AI model along with your message for analysis.

PDF Export

• Ask the AI to generate content in PDF format (e.g., "create a summary in PDF").
• A Download PDF button will appear below the response. Click it to save the file.
• You can also attach a PDF template and ask the AI to use the same format. The output PDF will match the template's header, footer, and layout.
• You can specify the number of pages (e.g., "create a 5-page PDF").

Prompt Library

• Click # Prompt Library in the sidebar to open it.
• Browse prompts by access level using the tabs: Organization, Department, Favorite, or My Prompt.
• Use the Module dropdown to filter prompts by module.
• Use the Search box to find prompts by title.
• Click any prompt card to preview it. From the preview, you can Run it directly or Open in New Chat.
• Click + Save Prompt to create a new prompt. Fill in the title, module, access level, prompt text, and optional tags.
• You can edit or delete prompts that you own.
• Mark any prompt as a favorite for quick access under the Favorite Prompt tab.

Usage Guide

• Click Usage Guide in the sidebar (bottom) to view this page at any time.

Sign Out

• Click Sign Out at the bottom of the sidebar to log out.
• All chat data will be cleared upon sign out.