Getting Started
Getting started with Hugo AI Agent
This article will guide you step-by-step to get your AI Agent setup and deployed in just minutes. Hugo is a next-generation AI support agent solution, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like a human support agent. Built to autonomously handle a large share of incoming conversations, Hugo can alsoFew readersHow to use Hugo Tools to set up and improve your AI Agent
Learn how to use Hugo Tools to create clearer instructions and workflows for your Hugo AI Agent. Hugo Tools help you move from a vague AI setup to clear behavior rules and workflow logic. They do not replace your Crisp configuration, but they help you prepare better inputs before applying them inside your workspace. In this guide, you will learn how to: Understand Hugo Tools → know what each helper is made for Use the Instruction Helper (1-use-the-instructFew readersHow to use Hugo Copilot to handle complex support conversations
This article explains how Hugo Copilot helps support agents handle complex conversations directly from the Crisp Inbox. Hugo Copilot is an internal AI assistant built for customer support operators. It helps your team understand customer requests, use available context, draft better answers, and prepare handovers while keeping the final response under human control. Unlike Hugo AI Agent, Copilot does not answer customers automatically. It suppoFew readersHow to migrate from MagicReply chatbots to Hugo AI Agent
This guide helps you migrate a legacy AI chatbot setup to Hugo while keeping the useful parts of your existing automations. Hugo is the recommended way to automate support with AI in Crisp. Instead of maintaining large workflow trees for every possible question, you can let Hugo answer from your knowledge, follow clear instructions, start workflows when a structured process is needed, and escalate safely when a human should take over. Legacy AIFew readersHow to navigate the Hugo interface and features
Feeling overwhelmed or lost? This article provides an overview of the features available with Hugo, and where to find them. A great place to start with or to come back when looking to improve your setup! Hugo is a next-generation AI support agent solution, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like aFew readersTroubleshooting & Common questions about Hugo AI Agent
This article helps you diagnose the most common Hugo issues without losing the useful context behind each answer. Hugo is designed to be easy to configure, but AI behavior can still depend on several layers at once: training resources, instructions, routing, escalation settings, model choice, integrations, and the conversation channel. Use this guide when Hugo gives an unexpected answer, does not answer, escalates too early, or behaves differently in production than in your tests. Haven't mFew readers
AI Agent - Guides
How does Hugo Routing work
Routing rules are used to describe situations you wish Hugo to detect. When he encounters these situations, he can then trigger the action of your choice, giving you flexibility in all kind of workflows. Hugo is a next-generation AI support agent, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like a human support agent. Built to autonomFew readersHow to start Hugo from Workflows
This article explains how to hand a workflow over to Hugo and when a workflow-first setup makes sense. Most teams get the best results by letting Hugo handle conversations first, then starting workflows only when a structured process is needed. However, starting with a workflow can still be useful when you need to collect mandatory information, apply strict conditions, or run a specific flow before Hugo answers. A workflow-first sSome readersHow to train Hugo AI Agent on your data
This article explains how to train Hugo with your own resources so it can answer customers accurately. Good training data is the foundation of Hugo's answer quality. Use AI Agent → Train to provide website content, knowledge base articles, files, and Questions & Answers snippets, then keep those sources clean, current, and focused on what customers actually ask. If you prefer a quick walkthrough, this video shows how to train yourSome readersHow to prompt Hugo AI Agent for Instructions
This article explains how Hugo instructions work and how to write prompts that shape the AI Agent's behavior clearly. Instructions define how Hugo should behave when answering users: tone, boundaries, escalation habits, formatting preferences, and other response rules. They should not replace training resources; factual knowledge belongs in AI Agent → Train. If you prefer a quick visual walkthrough, this video shows how to struFew readersHow to configure Escalation with Hugo AI Agent
This article will guide you step-by-step to get your AI Agent setup and deployed in just minutes. Hugo is a next-generation AI support agent, fully integrated with Crisp. While conversational at its core, the AI Agent for customer support goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like a human support agent. Built to autonomously handle a large share of incoming conversations, HugoFew readersHow to build MCP integrations with Hugo
MCP integrations allow AI Agents to do something that was, until now, reserved to humans: use tools. They let Hugo go beyond simply answering questions from training or documentation, and instead interact with real systems, fetch live data, and take action — just like a human agent would. Hugo is a next-generation AI support agent, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactivelFew readersHow to Classify Customer Support Topics with Hugo AI
This article explains how Hugo Topics work and how to use them to understand the conversations your AI Agent handles. Hugo is a next-generation AI support agent, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like a human support agent. Built to autonomously handle a large share of incoming conversations, Hugo can also perform real actFew readersHow to use Hugo Integrations
This article explains which native MCP integrations can be connected to Hugo and how they help your team automate support actions. Hugo is a next-generation AI support agent, fully integrated with Crisp. While conversational at its core, Hugo goes beyond answering questions. He understands customer intent and context to proactively guide users toward resolution — just like a human support agent. Built to autonomously handle a large share of incoming conversations, Hugo can also perFew readersHow to use Hugo Analytics to improve your AI Agent
This article explains how Hugo Analytics helps you monitor AI Agent activity, outcomes, and improvement opportunities. Analytics make it easier to understand what Hugo handles, where conversations are escalated, and which topics generate the most volume. Use these insights to improve training resources, instructions, routing, and the overall automation strategy. Hugo Analytics allow you can monitor two types of data in the AI AnFew readers
AI Agent - FAQ
How to create an AI Chatbot for Customer Service with Claude
Learn how to set up Hugo, the Crisp AI Agent, as a Claude-powered AI chatbot for customer support. Hugo combines a supported Anthropic Claude model with your support content, business instructions, routing rules, and human handoff. This is sometimes described as a Claude customer support chatbot, but Hugo manages the support experience while Claude powers language understanding and answer generation. Discover Hugo to see how the AI Agent works.Few readersHow to create an AI Chatbot for Customer Service with Gemini
Learn how to set up Hugo, the Crisp AI Agent, as a Google Gemini-powered AI chatbot for customer support. Hugo combines a supported Google Gemini model with your support content, business instructions, routing rules, and human handoff. This gives teams a fast and cost-efficient customer support chatbot while keeping the complete support experience inside Crisp. Discover Hugo to see how the AI Agent works.Few readersHow to create an AI Chatbot for Customer Service with GPT
Learn how to set up Hugo, the Crisp AI Agent, as a GPT-powered AI chatbot for customer support. Hugo combines a supported OpenAI GPT model with your support content, business instructions, routing rules, and human handoff. This is sometimes described as a ChatGPT customer support chatbot, but Hugo manages the customer support experience while GPT powers language understanding and answer generation. Discover Hugo to learn what the AI Agent can do.Few readersHow to create an AI Chatbot for Customer Service with Mistral
Learn how to set up Hugo, the Crisp AI Agent, as a Mistral-powered AI chatbot for customer support. Hugo combines a supported Mistral AI model with your support content, business instructions, routing rules, and human handoff. This gives businesses a natural, multilingual customer support chatbot powered by a European model provider. Discover Hugo to see how the AI Agent works.Few readersHow to choose and configure Hugo’s AI Model
Learn how to choose and configure Hugo’s AI model settings based on answer quality, cost, compliance, and reliability. Hugo supports several AI model families and processing options. These settings help you balance answer quality, speed, credit usage, hosting region, fallback behavior, and media processing for your customer support use case. In this guide, you wiFew readersDoes use Hugo use conversational data
This article explains whether Hugo uses your Crisp conversations as training data and what conversational context it can access. Hugo can use the context of the current conversation to answer properly, but it is not trained on all previous conversations from your inbox. This distinction matters for privacy, security, and the quality of the answers Hugo provides. No. Hugo is not trained on all conversations from your CrispFew readers
Workflows - Guides
Getting started with Workflows
This article explains what Crisp Workflows are, how they can be started, and how to build your first automation. Workflows are no-code automations made of connected blocks. They help you collect information, offer choices, route conversations, update conversation data, trigger actions, and hand structured processes to Hugo when needed. They work best when you need a predictable flow, while Hugo is better for open-ended support questions. Haven't met Hugo yet? Check out official resources (hSome readersTips & Tricks about Workflows
This article collects practical tips to help you build, organize, test, and troubleshoot Crisp Workflows more efficiently. Workflows can stay simple or become highly advanced depending on your use case. The tips below focus on the details that make them easier to maintain: block organization, branch priority, testing habits, reusable flows, and common mistakes to avoid. *There's multiple ways to build chatbots by using workflows. You'll find that mostFew readersUnderstanding and Mastering the Crisp Workflows
This technical reference explains how each Workflow Builder block works, with practical notes to help you build reliable automations. Workflows are created by inserting and configuring blocks inside the no-code Workflow Builder. By connecting these blocks, you can build flows with actions, conditions, events, AI handoffs, and internal operations for your inbox conversations.PopularHow to create an Out-of-Office workflow
This article explains how to create an out-of-office workflow that replies when your team is unavailable. An out-of-office responder can set expectations, collect useful context, and keep the conversation organized until your team returns. You can also combine it with Hugo or knowledge resources when users should receive helpful answers before an agent follows up. In this video, you’ll learn how to build an Out-of-Office floSome readersHow to start Workflows programmatically (without the user sending a message)
This guide explains how to start Crisp Workflows from your website or backend, even before the user sends a message. Programmatic triggers are useful when your product already knows what should happen next: opening a guided support flow, collecting contextual information, starting onboarding from a button, or launching a workflow after a product event. Crisp supports both Web Chat SDK triggers for frontend use cases and the Workflow API for backend use cases. color 0080ddSome readersHow to start Workflows manually
This article explains how agents can start workflows manually from an existing conversation. Manual starts are useful when an agent wants to trigger a prepared flow at the right moment, such as collecting missing information, sending a guided form, booking a demo, or continuing with a structured process without rebuilding the same steps by hand. Workflows can also be started manually from any conversations by yourself and youFew readersHow to start Workflows situationally with Hugo AI routing
This article explains how Hugo can start workflows automatically when a conversation matches a situation you define. Situational routing is often the cleanest way to combine Hugo and Workflows. Hugo handles the conversation naturally, then starts a workflow only when the user reaches a topic, intent, or situation that needs a structured process. If you went through our Getting started with Hugo guide, you might recallFew readersHow to start Workflows automatically for all incoming conversations
This article explains how to start workflows automatically when new users contact you. Automatic workflow starts are useful when you want every new conversation to follow a structured flow first, such as collecting information, offering choices, or routing requests. If Hugo is active, you can also decide whether Hugo should answer first or route users directly into a selected workflow. If you wish to tFew readersHow to update user information in their profile with Workflows
This article explains how to update a user's Crisp profile from information collected in a workflow. When a workflow asks for details such as a name, email address, phone number, or custom value, you can save that information directly into the conversation profile. This helps your team keep customer context accurate and reuse collected data later in the same workflow. Updating user information is generally done after haFew readersHow to create an AI "WISMO" chatbot with Crisp?
In this guide, you will learn how to create an AI-powered support chatbot with Crisp that automatically answers "Where is my order?" requests using external order data. No code is required in Crisp, but you'll need access to your order system or a webhook service. Just discovering Crisp AI Chatbot? Check out our Getting Started with the Crisp AI Chatbot aFew readers
Workflows - Examples
How to send embedded and external links with Workflows
This article explains how to share embedded and external links from workflow buttons. Links are useful when a workflow should send users to a product page, booking page, article, checkout, or external resource. Crisp lets you decide whether a link opens directly in the chat experience or in a new browser tab. Inside of your workflow, you can open the Block Drawer from the left to drag a new Action block "Send Message: Button PFew readersHow to ask users for information and memorize them with Workflows
This article explains how to ask users for information in a workflow and store their answers as memorized values. Memorized values let a workflow reuse what the user submitted later in the flow. This is useful for details such as an email address, nickname, order number, request type, or any other value you want to insert into messages, conditions, or profile updates. Inside of your workflow, you can open the Block Drawer fromFew readersHow do I build a list of buttons with Workflows
This article explains how to show users a list of buttons and branch your workflow based on their choice. Button pickers are one of the easiest ways to guide users through a structured flow. They help users choose a department, topic, product, or next step without typing, and they make your workflow easier to test because each choice can lead to a predictable branch. Inside of your workflow, you can open the Block Drawer from theFew readersHow to create product carousels with workflows
This article explains how to add a product or item carousel to a Crisp workflow message. Carousels are useful when you want to showcase multiple products, plans, resources, or options in a compact message. Each item can include a title, description, image, and button so users can continue to the right page directly from the conversation. Inside of your workflow, you can open the Block Drawer from the left to drag a new Action block "SeFew readersHow to build Drip campaigns with Workflows
This article explains how to use workflows, events, segments, and campaigns together to build a simple automated drip campaign. A drip campaign works best when your product already emits useful events. A workflow can push or react to those events, update the user's context, and help your team send the right campaign message at the right step of the customer journey. Let's see in action howFew readersCreating a chatbot for sales teams
This playbook explains how sales teams can use Crisp workflows and Hugo to qualify leads, book meetings, and guide prospects faster. Sales automations work best when they remove repetitive steps without blocking the conversation. Use workflows for structured qualification and routing, and use Hugo when prospects need natural answers about your product, pricing, or next steps. For the purpose of this salesFew readers
Workflows - FAQ
What are pattern matches and how do they work
This article explains how pattern matches work in Crisp Workflows and how to use them to detect user messages reliably. Pattern matches let a workflow react to specific words, phrases, or structures in a user message. They are especially useful with User Message Matches blocks, where the right wildcard setup determines whether your workflow detects only exact messages or broader variations. A pattern is a regularity in the world. ThFew readersHow to configure Workflows for specific channels
This article explains how to make a workflow run only on the conversation channels you choose. Crisp can receive messages from website chat, email, WhatsApp, Messenger, Instagram DM, and other channels. By configuring the Origins field in your first event block, you can keep one workflow channel-specific or make it intentionally omnichannel. This event is the first step you should always set when building a chatbot (https:Few readersHow does the Automated inbox work
This article explains how the Automated inbox works when Hugo handles conversations for your team. When Hugo is active, Crisp groups conversations handled by the AI Agent in a dedicated Automated inbox. This keeps human inboxes focused on conversations that need agent attention, while still letting your team review Hugo's work and step in whenever needed. The Automated inbox is a dedicated inbox where conversations handled by HFew readersHow to measure AI Chatbot ROI with Crisp?
Artificial Intelligence doesn’t just resolve tickets, it changes how much work reaches your team and how much that work costs. In Crisp, AI ROI is not a vague promise. You can measure it concretely through your Analytics dashboards and a simple AI performance calculator that turns deflection and time saved into real financial impact. This guide walks you through the essentials of measuring AI performance in Crisp. You'll learn which metrics truly matter for evaluating your chFew readers