Track food calories via Telegram with GPT-4 Vision and Google Sheets
This workflow enables users to submit food images to a Telegram bot, which uses OpenAI’s GPT-4 Vision to identify the item and estimate its caloric value. The results are stored in Google Sheets and sent back to the user. What it does: Triggers on a photo sent via Telegram. Acknowledges the user with a sticky note message. Downloads the image file securely using Telegram's API. Sends the image to GPT-4 Vision with a prompt: “Describe this food and estimate its calories.” Logs the GPT response to a Google Sheet (with timestamp). Replies to the user with the result (e.g., food name and estimated calories). Use cases: Personal food tracking Nutrition logging via chat Meal journaling for fitness or health Requirements: Telegram Bot Token (via credentials) OpenAI GPT-4 Vision access Google Sheets credential with access to the target sheet Notes: You can extend this template to calculate daily totals, categorize meals (breakfast/lunch/dinner), or even integrate with calorie goals. The sticky note node confirms receipt to enhance UX. Ideal for wellness apps, chat-based food journals, or AI-powered health bots.
By ShivaAI Telegram-to-WordPress publishing with GPT-4 & media handling
AI-powered automation that rewrites, enhances, and publishes Telegram RSS content directly to your WordPress news site, including image/video upload and category mapping. Who’s it for This workflow is designed for content creators, news publishers, and social media managers who share updates on Telegram and want to automatically republish them as formatted articles on WordPress. It’s ideal for news portals, agencies, or blogs that manage content across multiple channels. Features Multi-Channel Telegram RSS Monitoring Automatic Category Assignment Duplicate Prevention System (GUID Matching) Media Download + Type Detection (Image / Video) Full AI Article Rewriting (GPT-4O) -- Stage 1: Full rewrite with structure -- Stage 2: Context enhancement Media Handling Logic WordPress Publish How it works / What it does Fetches new posts from a Telegram channel feed (via RSS). Uses OpenAI to rewrite the text into a polished news-style article. Detects whether the content includes images or videos and downloads them. Uploads the media to WordPress and links it to the post. Automatically publishes the formatted article to WordPress with the correct category and excerpt. Set up steps Setup takes around 10–15 minutes. You’ll need API keys for OpenAI and WordPress Application Passwords. Add your Telegram RSS feed URL and WordPress site URL in the relevant nodes. (Optional) Adjust tone or rewrite style in the OpenAI node and category mapping in the Switch node. All configuration details are included in sticky notes inside the workflow. Requirements WordPress site with REST API access and Application Password OpenAI API key Telegram channels RSS URL How to customize the workflow You can easily adjust the writing style in the OpenAI node, change categories in the Switch node, or schedule how often the workflow checks Telegram for new posts.
By Alexandru Burca