Quick overview This demo workflow listens for Telegram rental messages, routes receipt-related commands, uses OpenAI to classify payments, costs, and booking summaries, then replies with parsed details and a sample receipt flow. How it works Triggers when a new Telegram message is received. Routes the message into one of four paths: full /receipt command, receipt request, receipt date reply, or general rental tracking. If a full /receipt command is provided, it parses client, property, stay dates, amount, and transaction details and replies in Telegram with the extracted fields. If the user sends a receipt request, it looks up the last saved payment for that Telegram chat and either prompts the user for stay dates or returns an error if no payment is stored. If the user replies with stay dates during a pending receipt flow, it combines those dates with the stored payment to build receipt details and replies with either a “ready to generate” summary or an error message. For all other messages, it sends the text to OpenAI to classify it as payment, cost, or booking summary, normalizes the JSON result, and replies in Telegram with the parsed output.
Tags
TelegramRelated workflows
See all AI Automation→
C
WExtract and log Gmail receipt expenses with OpenAI, Sheets and Airtable
Quick overview This workflow monitors a Gmail “Receipts” label, downloads attachments, uses OpenAI to extract expense data from receipt emails and images, skips duplicates and non-receipts, fails safely when extraction is unreadable, sends Slack/Teams alerts for high-value, warranty, or review-needed receipts, and logs valid receipts to Sheets or Airtable. How it works Triggers every minute when Gmail finds new emails matching the query label:Receipts newer_than:2d. Normalizes the email into consistent fields (message ID, subject, sender, received date, plain-text body) and generates a receipt hash for stable matching. Searches Airtable and/or Google Sheets for existing records matching the message ID or receipt hash and marks the email as a duplicate if it falls within the configured duplicate window. For new receipts, sends the email content to OpenAI to extract receipt details like merchant, total, category, line items, tax hints, and return/warranty dates. Evaluates the extraction to flag high-value, warranty/return-relevant, or needs-review receipts and formats alert messages. Posts alerts to Slack and/or Microsoft Teams when enabled and the receipt is not “standard.” Appends
CWExtract and log Gmail admin tasks with OpenAI, Google Sheets and Airtable
Quick overview This workflow monitors Gmail for messages labeled “Admin”, redacts sensitive data, uses OpenAI to extract structured admin tasks and reminders, skips duplicates by checking Google Sheets and/or Airtable, sends optional Slack or Microsoft Teams alerts for urgent items, and logs the clean task details. How it works Triggers every minute when a new Gmail message matches the query label:Admin newer_than:7d. Loads configurable settings (log destination, alert channels, due-soon window, dedupe window, and redaction toggle) and normalizes the email into a clean text body. Redacts sensitive patterns (such as SSNs, card-like numbers, phone numbers, email addresses, and addresses) and generates an “admin hash” fingerprint for deduplication. Checks Google Sheets and/or Airtable for recent matching Message IDs or admin hashes and skips processing if the email appears to be a duplicate. Sends the redacted email content to OpenAI to classify whether it is an admin item and to extract fields like task title, summary, priority, due date/time, and recommended next step. If the item is valid, optionally posts an alert to Slack and/or Microsoft Teams for urgent/high-priority, due-soon,
WGenerate weekly Pinterest content opportunity briefs with Apify, OpenAI and Google Sheets
Quick overview Research a Pinterest niche with FetchCat's Pinterest Search Scraper, preserve every source pin in Google Sheets, and use one evidence-controlled OpenAI call to identify recurring themes, underrepresented angles, and five production-ready content briefs without claiming search volume or trend growth. How it works Starts manually with one research niche and one to five exact Pinterest search phrases, plus locale, country, and a configurable result limit. Runs fetch_cat/pinterest-search-scraper through a Cloud-compatible HTTP Request and collects 20 to 500 public pins per search with public detail enrichment enabled. Validates, normalizes, and deduplicates returned pins. It stops before AI or destination writes when fewer than ten usable pins remain. Builds stable Google Sheets rows, deterministic phrase counts, and a bounded evidence packet, then uses one structured gpt-5.4-mini request for analysis. Removes invalid pin citations and rejects any finding that retains no supplied evidence, preventing unsupported recommendations from reaching the destination. Upserts source pins and a readable brief into separate Pins and Research Brief tabs using stable same-day research