Basil Irfan
Templates by Basil Irfan
Automate restaurant reservations with AI on WhatsApp and Google Sheets
Streamline restaurant reservations on WhatsApp Overview This n8n template automates table bookings via WhatsApp, letting users request, confirm, and manage reservations without manual intervention. It leverages AI to parse messages, apply group discounts, check availability, and send natural confirmations—all within a single, reusable workflow. Key Features AI‑powered parsing & responses: Extracts guest name, date, time, and party size from free‑form WhatsApp messages and generates friendly confirmations.. Availability lookup: Integrates with Google Sheets, Airtable, or MySQL to verify slot availability in real time. Automated reminders: Optionally schedules follow‑up messages 24 hours before the booking. Modular design: Swap triggers, storage, or messaging nodes to fit your infrastructure. How It Works Trigger: Incoming WhatsApp message via WhatsApp Business Cloud API. Parse & Validate: AI Function node extracts intent and guest details. Calculate Discount: Custom Function node computes group discount. Compose Confirmation: Open Ai text model generates a personalized response. Send Message:Request node posts back to WhatsApp. Optional Reminder: Wait node + HTTP Request for pre‑booking follow‑up. Requirements WhatsApp Business Cloud API access n8n Cloud or self‑hosted instance Reservation datastore (Google Sheets, Airtable, MySQL) Open ai key for AI text generation Customization Tips Menu Attachments: Add media nodes to send PDFs or images. Alternate Slot Suggestions: Use AI to propose new times if a slot is full. Upsell Offers: Follow up with add‑on suggestions (e.g., wine pairings). Localization: Extend prompts for multilingual support.
By Basil IrfanLinkedIn lead generation with GPT-4o, Apify scraping, and automated outreach
🚀 LinkedIn Lead-Gen Flywheel – Apify → GPT-4o → Google Sheets → Phantombuster What this workflow does Collect audience specs – simple web-form asks for your ideal company profile. Generate a laser-targeted Apollo search URL with GPT-4o (no manual filtering). Scrape the matching leads via an Apify actor (returns clean JSON). Craft hyper-personalized icebreakers for each lead using GPT-4o (ultra-short, human-sounding). Log everything to Google Sheets – name, LinkedIn URL, company site, summary, and the icebreaker. (Optional) Auto-launch Phantombuster to fire off those connection requests at scale. Why it matters Zero grunt work: audience research, scraping, copy-writing, and outreach all run hands-free. Punchy personalization: micro-icebreakers outperform canned intros, boosting accept rates. Scales with you: flip a switch to go from 10 to 1 000+ connections/day. Node rundown | Step | Node | Key Inputs | Key Outputs | |------|------|-----------|-------------| | 1 | Form Trigger | Audience description | descriptionofcompany | | 2 | OpenAI (GPT-4o) | Audience text | SearchUrl | | 3 | HTTP Request – Apify | SearchUrl, APIFY_TOKEN | Lead JSON | | 4 | OpenAI (GPT-4o) | Lead JSON | Icebreaker | | 5 | Google Sheets | Lead + Icebreaker | Row append/update | | 6 | Aggregate | Sheet rows | Batched output | | 7 | HTTP Request – Phantombuster | PHANTOMKEY, AGENTID | Launch status | Prerequisites OpenAI API key (GPT-4o access recommended) Apify API token with access to actor id Google Service Account creds shared with your target sheet Phantombuster API key and Agent ID for your LinkedIn connector Active Apollo account to open the generated search URL (only required for debugging) Setup (5-minute sprint) Import the workflow into n8n. Add the required credentials in Credentials → OpenAI, Apify, Google Sheets, Phantombuster. Paste your Phantombuster Agent ID into the HTTP Request node URL. Publish the Form Trigger URL—this is where you (or your SDRs) describe the target audience. Hit Execute Workflow once to verify data flows end-to-end. Customization tips Titles & keywords: tweak the prompt in the first GPT-4o node to lock in different roles or industries. Icebreaker style: adjust the second GPT-4o prompt to match your brand voice. Data columns: map extra fields from Apify into Google Sheets as needed. Skip outreach: disable the Phantombuster node if you only want the leads + icebreakers.
By Basil IrfanTransform YouTube videos to social media content with Vizard AI and GPT-4.1
Transform YouTube Videos to Social Media Content with Vizard AI and GPT‑4.1 Overview This n8n template fetches new YouTube videos, enriches them with Vizard AI metadata, generates social‑media captions using GPT‑4.1, logs everything to Google Sheets, and notifies you by email. It’s a turnkey solution for content creators and marketers who need an end‑to‑end automated pipeline from video publishing to post scheduling. Setup Instructions Import the Template In n8n, click Import from JSON, paste this workflow, and save. Configure Credentials Vizard AI: Create an HTTP Request credential named Vizard API and set your VIZARDAIAPIKEY. OpenAI: Add a new OpenAI credential for GPT‑4.1. Google Sheets: Create a Google Sheets OAuth2 credential with read/write access or just sign in if your on cloud hosting Gmail: Add a Gmail OAuth2 credential for email notifications or just sign in if you are on cloud hosting Adjust Limits In the Limit Videos node, set maxItems to control batch size. Google Sheets Column Structure | Column | Description | | ------------------ | ---------------------------------------------------- | | videoId | Unique YouTube video identifier | | projectId | Vizard AI project ID returned | | videoUrl | Original YouTube video URL | | title | Video title | | transcript | Transcribed text from Vizard AI | | viralScore | Vizard AI’s viral‑score metric | | viralReason | Explanation for viral score | | generatedCaption | GPT‑4.1–generated caption in JSON { "caption": ""} | | clipEditorUrl | URL to Vizard’s clip editor | Workflow Steps Read YouTube RSS Feed (Read YouTube RSS Feed) Limit Videos (Limit Videos to N) Send to Vizard (Create Vizard Project & Retrieve Vizard Metadata) Split Items for Processing (Iterate Each Video) Generate Captions (Generate Social Media Captions) Append Row in Sheet (Log to Google Sheets) Send Notification (Email Summary) Customization Tips Alternate Caption Styles: Modify the AI prompt for tone, length, or brand voice. Localization: Extend prompts for other languages. Notification Channels: Swap Gmail for Slack, Teams, or SMS via webhook nodes.
By Basil IrfanWhatsApp support bot with Google Drive RAG, GPT-4.1-mini and Cohere reranking
WhatsApp RAG Agent (Text + Voice) with Weekly Google Drive Sync One-line summary : Answers WhatsApp in under 100 words, understands voice notes, and retrieves trusted answers from your Google Drive docs (RAG) kept fresh weekly. --- What this template does Reply to WhatsApp messages in a polite, human tone with ≤100 words. Understands text and voice notes: auto-downloads audio and transcribes to text. Retrieves answers from your knowledge base (RAG): Google Drive docs → chunk → embed → store in Supabase → rerank with Cohere. Keeps short-term memory across the conversation to avoid repetition. Weekly doc sync from a selected Google Drive folder so non-technical staff can update content without touching n8n. --- Why it matters Zero busywork: Update a Google Doc; the bot learns it on the next sync. Trustworthy answers: Responses come from your vetted docs, not random web text. Voice-first friendly: Handles the WhatsApp reality of “send a voice note.” Safer by design: Guardrails—no pricing unless in KB, no pushy sales, no medical advice. --- Triggers WhatsApp Trigger: Receives incoming messages (text or audio). Google Drive Trigger (weekly): Detects new/updated files in the chosen folder for ingestion. --- App credentials required WhatsApp Business Cloud (App ID, Token, Phone Number ID) OpenAI (Chat + Embeddings) Cohere (Rerank) Supabase (SUPABASEURL, ANONKEY) Google Drive (OAuth2) + target Folder ID --- Suggested environment variables bash WHATSAPPAPPID= WHATSAPP_TOKEN= WHATSAPPPHONENUMBER_ID= OPENAIAPIKEY= COHEREAPIKEY= SUPABASE_URL= SUPABASEANONKEY= GDRIVEFOLDERID= RAGTABLENAME="documents" table to store vectors/metadata MAXANSWERWORDS=100 guardrail for concise replies --- Architecture overview Answer-time lane (RAG Tool): Receive WhatsApp message → 2) Transcribe audio if present → 3) Maintain short-term memory → 4) Retrieve from Supabase vectors (topK) → 5) Rerank with Cohere → 6) Compose ≤100-word reply with sources → 7) Send on WhatsApp. Ingest lane (Weekly Sync): A) Detect Drive file updates → B) Download & convert (Docs → text/plain) → C) Chunk (size/overlap) → D) Embed with OpenAI → E) Upsert to Supabase with metadata & hashes. --- How it works (node rundown) | | Node | Key Inputs | Key Outputs | | -- | ------------------------------------------- | ------------------------------- | --------------------------------- | | 1 | WhatsApp Trigger | Incoming message | Raw WhatsApp payload | | 2 | Switch (Attachment presence/type) | Payload | Route: Text or Audio | | 3 | HTTP Request (Audio path) | attachments[0].data_url | Audio file | | 4 | OpenAI – Translate/ASR | Audio file | Transcribed text | | 5 | Merge | Text path + Audio path | Unified text message | | 6 | Simple Memory | Recent turns | Short-term context | | 7 | OpenAI Chat Model | Prompt + message + memory | Draft answer (tool calls allowed) | | 8 | Supabase Vector Tool (retrieve-as-tool) | Query text, topK=10 | Candidate KB passages | | 9 | Cohere Reranker | Candidates | Re-ranked context | | 10 | Send WhatsApp Message | to, body | Reply sent | | 11 | Google Drive Trigger (weekly) | Folder ID, fileUpdated | Changed files | | 12 | Set (File Id) | id from trigger | File ref | | 13 | Google Drive – Download File | Id (Docs→txt) | Raw text | | 14 | Character Text Splitter | chunkSize=2000, overlap=300 | Chunks | | 15 | Default Data Loader | Binary→Document | Clean docs | | 16 | OpenAI Embeddings (ingest) | Chunks | Vectors | | 17 | Supabase Vector Store (insert) | Table: documents | Upserted KB | Notes The KB Tool is the combo of steps 8–9–7 at answer time (retrieve → rerank → answer). The Ingest lane is steps 11–17 (weekly sync of your Drive folder). --- Setup (7‑minute sprint) Import the workflow JSON. Connect credentials: WhatsApp, OpenAI, Cohere, Supabase, Google Drive. Google Drive Trigger: paste your Folder ID; keep fileUpdated event. Download File: ensure Google Docs convert to text/plain. Supabase Vector Store (insert): set table name to documents (or your schema). Character Text Splitter: keep chunkSize=2000, overlap=300 (balanced recall/latency). Retrieve-as-tool: set topK=10 and enable reranker. Send WhatsApp Message mapping: Recipient: {{$("WhatsApp Trigger").item.json.messages[0].from}} Body: {{$json.output}} Test: Send a text and a voice note to your WhatsApp number → confirm concise answers. Drop a Google Doc into the watched folder → verify it’s chunked/embedded on the next weekly poll (or run ingest nodes once manually). --- Prompt, tone & guardrails System prompt: Be polite, human, and concise (≤ MAXANSWERWORDS). Cite or reference only the KB content; if unknown, say so and offer to escalate. No prices unless present in the KB. No medical advice. Temperature: start at 0.2 for factual replies. Memory window: keep a short rolling buffer (e.g., last 4–6 turns). --- Data model (minimum viable) Table documents (example columns): id (uuid) source_url (text) title (text) chunk (text) embedding (vector/float[] depending on extension) chunk_hash (text) updated_at (timestamp) Indexes Unique index on chunk_hash to dedupe Index on updated_at for syncs --- Observability & ops Log question, selected chunk ids/hashes, and final response to a DB/Sheet for QA. Add a low-confidence route (score threshold) → Slack/Telegram escalation to a human. Track latency and token usage to tune topK and chunk sizes. --- Customization Latency vs quality: try topK=6–8 and chunkSize=1200 for speed. Languages: ASR node can be swapped for native multilingual output. Escalation: add channel handoff on low confidence or no KB hits. Sync cadence: change Drive Trigger to daily if content updates frequently. --- Safety & compliance No medical advice. If uncertain or clinical, ask to schedule a consult or refer to the right department. PII: Don’t log full phone numbers in plaintext analytics; hash where possible. Prices & rates: Only answer if present in the KB; otherwise hand off to front desk. --- Troubleshooting No reply sent: Ensure Send message node reads {{$json.output}} (Agent’s response property). Audio path failing: Confirm attachments[0].data_url exists and HTTP node fetches a valid file. KB not updating: Manually execute the ingest lane; check rows in Supabase. Irrelevant answers: Lower temperature to 0.2, increase overlap to 400, and verify Drive docs are clean and structured. --- Categories & tags Categories: AI, Customer Support, Healthcare Ops, RAG, WhatsApp Tags: WhatsApp, RAG, Google Drive, Supabase, OpenAI, Cohere, Voice Notes --- Pricing (rough, BYO keys) n8n: self-host free; n8n.cloud billed by plan. OpenAI, Cohere: usage-based by tokens/calls. Supabase: free tier + usage; vector storage billed by size. WhatsApp Cloud: Meta pricing per conversation. --- Nodes used in workflow WhatsApp Trigger, Switch, HTTP Request, OpenAI (ASR + Chat + Embeddings), Merge, Simple Memory, Supabase Vector Tool (retrieve), Cohere Reranker, Google Drive Trigger, Set, Google Drive – Download, Character Text Splitter, Default Data Loader.
By Basil Irfan