An n8n agent searches a knowledge base and order history in parallel, uses GPT-4o-mini to classify and draft replies, emails resolvable tickets, and routes billing/refund/security cases to Slack with a draft (~$0.90/mo at 50 tickets/day).
n8n’s 2026 agent examples show a Sheet of URLs → AI visits each site → tailored outreach draft + lead score, so SDRs stop pasting the same cold template.
n8n’s starter template wires Chat Trigger, an AI Agent node, Gemini, weather + RSS tools, and short memory—so you prove tool-using agents before building the fancy multi-agent graph.
n8n docs describe first-class agents with tools, knowledge bases, hourly/daily schedules, and sub-agent handoffs—plus embedding agents inside fixed workflows via the Message an Agent node.
Official n8n AI example: a chat agent answers with GPT-4; if unsure it calls a sub-workflow that asks for the user's email and posts the unanswered question to a Slack support channel.
This n8n template extracts invoice PDFs with GPT-4o, checks invoice numbers and totals, assesses category and risk, and uses a wait-and-resume approval path before recording the final status.
An n8n template combines an Ollama agent, three Supabase vector knowledge bases, Postgres memory, request validation, deduplication, and confidence-based escalation across Slack, Gmail, and callbacks.
This n8n workflow finds completed online meetings, retrieves their transcripts, and lets an AI agent identify follow-up meetings and assemble Calendar events with known attendees.
Dataki's introductory n8n template gives an OpenAI-backed agent just two calendar capabilities: retrieve events and create an event. Sticky notes explain the setup inside the workflow.
An n8n template separates menu costing, leftover recipe suggestions, a nutrition diary, and draft social copy into specialist agents writing to dedicated Google Sheets tabs.
This n8n community template transcribes Telegram voice messages, extracts meal ingredients and estimated nutrients with OpenAI, stores structured rows in Sheets, and replies with a confirmation.
A builder reports that a two-court booking system became more reliable after moving availability checks, resource assignment, and booking IDs out of the LLM and into deterministic n8n workflows.
Local-first OSS DB/workspace where Claude Code/Codex/Cursor share durable records, docs, skills, and apps—writes that matter go through human review; Skill, MCP, OpenAPI, and ACP connections keep agents stateless and work durable.
This workflow records agent outcomes in an n8n Data Table with linked SHA-256 hashes, provides a verification endpoint, and prepares daily integrity reports.
The template rejects OTPs and unknown senders, structures transaction alerts into a Sheets ledger or review queue, and answers spending questions through a Claude Telegram agent.
The workflow searches a live product API, waits separately for selection and purchase approval, then submits a checkout request and reports the asynchronous order result in Slack.
OpenAI categorizes submitted expenses and marks those needing approval, while the workflow records status in Sheets and sends managers details through Telegram and Gmail.
The workflow takes a room photo and property facts, calls staging and comparable-listing APIs, and uses Claude to prepare a draft listing package for an agent.
The workflow scores incoming mail for prompt-injection patterns, routes high-risk messages to review, and uses Gemini for structured triage of the remaining messages.
OpenAI vision extracts receipt totals and currency; deterministic checks compare them with expected booking values and route uncertain or mismatched documents to Slack.
GPT-4o-mini summarizes resume PDFs linked from a candidate sheet, emails the interviewer a structured brief, and marks the row processed to avoid repeats.
An Indie Hackers walkthrough connects a Jotform request, n8n planning and content generation, a GitHub branch, a Vercel preview, an approval step, and post-change monitoring.
WayinVideo summarizes a course video and GPT-4o-mini turns the learning points into structured tasks with estimates and checklists, which are created in ClickUp and logged in Sheets.
GPT-4o extracts property details from configured Facebook-group posts and classifies them against a defined buying brief, with qualifying links delivered through Twilio.
A Firecrawl-assisted template searches multiple job boards and uses an OpenAI agent to score listings against salary, location, stack, and other stated requirements.
The template computes property metrics from price, rent, and expenses, checks duplicate records, and uses Groq-hosted AI to prepare an explanatory email report.
The template runs question-and-reference-answer cases through a support agent and uses a separate GPT-4 judge to score open-ended responses in n8n's evaluation view.
The Telegram template searches Amazon, Walmart, and Google Shopping in parallel, normalizes offers, and uses OpenAI to assess variant matching and suspicious listings.
Azure OpenAI compares buyer preferences with a property database, retrieves relevant Drive decks, and helps organize the lead record and follow-up priorities.
The Gemini workflow reads product-image URLs, generates image prompts, combines products with a supplied model image, and saves images and prompt records to Drive and Sheets.
The template combines GPT intent detection, MongoDB product search, Redis sessions, and Gemini image generation for a conversational shirt-shopping flow.
The workflow selects a current article from RSS feeds, drafts a Telegram summary with Gemini, and waits for a human approval action before publishing it.
OCR and AI parse attached timesheets into month-specific billable hours, including weeks spanning months, before the workflow creates QuickBooks invoices and an audit sheet.
Gemini transcribes and analyzes a submitted video to produce Cornell notes, Feynman-style explanations, micro-goals, and practice activities in Notion.
The template reads PDF or DOCX resumes from a Notion candidate database, compares them with the linked job requirements using OpenAI, and writes scores and feedback back to Notion.
OpenAI assesses CRM, support, and PostgreSQL usage data; the workflow can create a Stripe coupon and compose a tailored retention email for selected customers.
An AI classifier turns school emails into calendar events, Drive notes, and categorized notices, including reminders for the following day's activities.
A GPT-4.1 assistant transcribes voice or reads text, answers product questions from store information, and records order details in Google Sheets while retaining conversation context.
An AI agent checks a Sheets list of birthdays and anniversaries, uses relationship notes to personalize messages, and passes matching greetings to an email step.
An AI workflow subtracts travel time to the next meeting from a calendar gap, then recommends a nearby cafe matching preferences or sends a departure reminder.
This workflow OCRs uploaded receipts, stages parsed blocks in Supabase, and uses a travel-expense agent with a calculator to return a structured reimbursement breakdown.
The template connects a locally hosted Signal API bridge to an OpenAI-powered n8n assistant, allowing additional tools to be added to the chat workflow.
Claude returns sentiment, tone, reply, confidence, and handover fields; post-processing applies length and content rules before routing a response for review.
The published workflow gathers conference details, exchange rates, and flight quotes, then prepares a budget-aware approval email and PDF expense breakdown.
An iOS Shortcut uses on-device Voice Memos transcription, optionally summarizes the text, and sends it through n8n as Markdown to a Drive-synced Obsidian folder.
An English video walkthrough connects n8n, OpenAI, Gmail, Calendar, and research tools to a personal assistant, demonstrating the construction of an integrated assistant rather than reporting a controlled benchmark.
The workflow combines HubSpot records, support context, and usage data from Sheets, then uses AI to assess customer health and explain recommended next steps.
An English n8n tutorial uses Google Vision OCR for PDF and image invoices, an LLM for structured fields, and Google Sheets for storage, making the OCR-to-extraction sequence explicit.
Claude converts the day's Google Calendar events into a concise message that Twilio sends by SMS, allowing a calendar briefing without opening another app.
The workflow retrieves recent AI papers, uses Gemini to assess relevance, and emails structured summaries of publications that pass the user's criteria.
The workflow reads survey data from Sheets, groups responses by sentiment, analyzes batches with AI, and saves a consolidated thematic report with Slack notification.
After creating QuickBooks invoices from Jotform orders, this workflow uses an AI agent to decide reminder actions and sends selected follow-ups through Outlook.
The template indexes provided learning documents in MongoDB Atlas and uses a Gemini Telegram agent to retrieve relevant passages before answering questions.
An OpenRouter-backed agent reads open Todoist tasks, prioritizes them using configured criteria, and posts a concise plan with overdue-item warnings to Slack.
A Telegram workflow uses vision to identify ingredients and a second model step to generate three structured recipes, then formats the suggestions for chat.
A Jotform-triggered workflow combines OCR and Gemini to calculate a receipt-based customer reward, checks the result, and logs the transaction in Notion.
An AI workflow retrieves meeting minutes from Drive, analyzes communication and participation, and returns a structured feedback report with suggested improvements.
Gemini drafts a reply using the sender's HubSpot contact, company, deal, and ticket context; the workflow sends through Gmail only after Slack approval.
Mistral OCR reads bank statements and GPT-4 proposes transaction-to-invoice matches, returning a reconciliation table with unmatched and review-needed items.
Gemini summarizes selected news while parallel branches retrieve weather and calendar events; Google speech synthesis and FFmpeg turn the combined script into audio.
OCR and Claude identify invoice and payment data; a Slack reviewer approves partial or final payment updates, with receipt records related back to each Notion invoice.
This self-hosted template combines a resume and job-description PDF, uses Perplexity to produce a tailored HTML resume, and converts it to a PDF stored in Drive.
The template validates incoming orders, checks local calling hours, and uses a Vapi voice agent to confirm order details while updating status records.
Gemini summarizes new article or YouTube links from a Sheets queue, stores the results, and lets a Telegram assistant retrieve the saved reference material.
The Gemini-powered backend receives a resume and job description, retains interview history, and produces follow-up questions for a voice-enabled web frontend.
A self-hosted workflow parses a travel query with Gemini, obtains accommodation listings through MCP, and combines them with live Amadeus flight offers.
This workflow combines public company research with specialist AI analyses of sustainability and animal welfare, producing a structured assessment and narrative evidence.
The template combines transaction details, IP geolocation, and Gemini analysis into booking-risk categories, with a Google Sheets audit trail and configurable responses.
The template retrieves a TikTok transcript, uses GPT-4 to identify pain points and triggers, and drafts a product-related post recorded alongside the research.
Dumpling AI retrieves a business's reviews and GPT-4 identifies recurring pain points, motivations, and voice-of-customer material in a Google Sheets research log.
This self-hosted template receives an Apify job dataset, filters relevant roles with AI, researches company decision-makers through Tavily, and updates Google Sheets.
The smart-calendar template combines OpenAI analysis, Perplexity research, and weather data to prepare contextual event notes delivered through Telegram.
Gemini analyzes three restaurant data streams in Sheets and merges sales highlights, waste-reduction suggestions, and customer-feedback themes into an email report draft.
The author's gift project combines a birthday webpage with an AI character that reads preferences and stores the evolving fictional story state in PostgreSQL.
The workflow reads language, proficiency, and interests from Sheets, researches current stories through Perplexity, and uses DeepSeek to produce level-adjusted reading, vocabulary, and grammar notes.
This workflow processes unread messages carrying a receipt label, extracts financial fields with OpenAI, stores attachments in Drive, and marks processed messages as read.
A self-hosted n8n template uses Mistral to structure chat-submitted expenses, archives receipt images in Drive, and links the resulting Google Sheets rows to the originals.
A VLM Run workflow parses emailed resumes, analyzes linked GitHub activity, and records a technical candidate profile with notifications for the recruiting team.
The published agent workflow alternates idea generation and critical review, then uses specialist prompts to develop an accepted campaign's audience, tactics, budget, and measures.
A Gmail workflow uses GPT-4 to classify messages and summarize important content, with Telegram alerts for items that meet the configured escalation criteria.
This language-learning workflow combines translation, AI-generated pronunciation and example sentences, and image retrieval to build vocabulary material suitable for flashcards.
An OpenAI email agent combines Outlook messages with contacts synced from Monday.com and classification rules held in Airtable to set categories and importance.
The demonstration retrieves portraits from Drive and uses Gemini vision plus supplied passport-photo requirements to return a structured validity assessment.
An AI workflow analyzes product feedback, considers whether a coupon is appropriate for negative sentiment, and produces a personalized customer email template.