> For the complete documentation index, see [llms.txt](https://docs.fullwhere.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.fullwhere.com/documentation/documentation-en/automations/ai-agents.md). # AI agents {% hint style="success" icon="list-check" %} **IN SHORT** Fullwhere AI agents let you set up automatic response generation. They can be customized using different instructions and variables. {% endhint %} {% hint style="info" icon="lightbulb" %} You can configure different AI agents depending on the context in which you want to use them.\ For example, you can choose to specialize your AI agents by interaction type, by brand, or by location. Once your AI agents are configured, use the [Workflow](/documentation/documentation-en/automations/workflows.md#generer-une-reponse) to define their scope and use. {% endhint %} {% hint style="info" icon="lightbulb" %} If your AI agent requires the use of in-depth knowledge to respond appropriately to your customers, it may be necessary to configure a [knowledge base](/documentation/documentation-en/automations/knowledge-bases.md) and use it with your AI agent. {% endhint %} {% hint style="info" icon="shield-check" %} **ACCESS RIGHTS** To edit the AI Agents in your Fullwhere workspace, you must be a supervising member with Administrator rights. {% endhint %} {% hint style="warning" %} You won't be able to delete an AI agent as long as it is used in at least one of your workflows. {% endhint %} ## Access your agents Go to the Automation → menu [Agents](https://app.fullwhere.com/automation/agents) from the main navigation bar on the left of your screen.\ You will find the list of all your existing agents, with several key pieces of information: * Agent name * Description (optional) * Created by: user who originally created it * Creation and last modification date * Workflows: number of workflows in which the agent is used To create a new agent, click + Create an agent. *** ## Create an agent To create a new agent, click + Create an agent from the main agents page.\ Enter your agent’s name in the window provided. ## Agent configuration Your AI agent is configured through 2 tabs: * Writing * Custom responses ### 1. Tab *Writing* This tab defines your AI agent's writing behavior. * Agent objective: this field is intended to specify the role your AI agent should fulfill (*e.g.: You must generate a response to answer my customers who contact me on ...*). * Writing instructions: these allow you to indicate to your agents your brand's tone of voice and how it should address your customers.\ ​*Examples:* * The way to say "hello" * Phrases to use or not use * The level of language to use (informal, polite, fun) * The way to address customers (using tu or vous), define the tone, phrases to use or avoid. * Sign-offs or polite closings to use * Use of emojis * etc. * Customer context: provides your AI agents with information about the current context when it has to generate a response for a customer interaction, such as the customer's first name, the location concerned, the communication channel, the customer's status, etc. * Additional instructions: allow you to specify any other relevant guidance. In these different configuration fields, it is possible to use dynamic variables. These variables will be replaced by the data from each customer interaction when generating a response.\ ​\ ​[View variables](https://help.fullwhere.com/fr/articles/12788791-les-agents-ia#h_3198d4840a). *** ### 2. Tab *Custom responses* The Custom responses tab lets you create response templates that the AI agent can rely on to generate messages that are more precise and consistent with your brand tone. These templates serve two main purposes: 1. Provide a smart generation base:\ when the AI detects a customer interaction (review, message, complaint...) matching one of your templates, it uses it as inspiration to formulate a response with the same meaning, while being able to adapt the wording or sentence structure. 2. Shape the brand tone:\ custom responses also serve as a style guide. They help the agent better understand how your brand expresses itself, the wording used, the level of formality, or even the use of emojis. Custom responses are organized into Groups, each corresponding to a theme (for example: “Positive reviews”, “Complaint”, “Cleanliness issue”, etc.). Each group contains several response examples with: * a label (detection criterion, e.g.: “5-star rating — no text”) * a sample response text, which you can freely structure with [dynamic variables](/documentation/documentation-en/automations/ai-agents.md#h_3198d4840a). ### Create a custom responses group From the Custom responses tab, click + Create a group to open the configuration panel and enter: * Application criteria: context in which the AI should use these responses (e.g. positive reviews, delivery complaints). * Custom response examples: add one or more sample responses depending on the situation, specifying each response's Label and Text. The text of custom responses can be built with dynamic variables. These variables will be replaced by the data from each customer interaction when generating a response.\ ​\ ​[View variables](https://help.fullwhere.com/fr/articles/12788791-les-agents-ia#h_3198d4840a). Click Create group to confirm. ### Manage your custom responses Once your custom response groups and responses are created, you can manage them from the custom responses table in the Custom responses tab. From this table, several actions are possible: * Edit a custom response directly:\ The label and text of each response can be edited directly in the table rows. * Open the full editing page:\ By clicking the Open button on a custom response row, you access the detailed editing page for that response. * Rename a response group:\ Group names can also be edited directly from the table. *** ## Use dynamic variables Writing instructions and custom responses can be built using dynamic variables.\ These variables are replaced with data from the relevant customer interaction when generating a response. Some variables will not have a replacement value depending on the type of customer interaction being handled: for example, an email will not have a replacement value for the variable `{{rating}}`. In this case, the variable will be replaced by an empty text string. ### General variables | Variable | Description | | ------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `{{tags}}` | List of feedback tags, displayed in brackets and separated by commas. | | `{{content}}` | Text content of the customer feedback. | | `{{rating}}` | Rating given by the customer in the feedback. | | `{{language}}` | Target language of the feedback. | | `{{source}}` | Source of the feedback (e.g.: Google, Email, Instagram, etc.). | | `{{class}}` | Class or category of the feedback (Review, Contact, Survey, Message) | | `{{store}}` | Name of the relevant location (e.g.: store, restaurant, branch…). | | `{{store_signature}}` | Signature associated with the location. | | `{{store_contact}}` | Location contact email address. | | `{{brand}}` | Brand name associated with the response. | | `{{brand_signature}}` | Brand signature. | | `{{brand_contact}}` | Brand contact email address. | | `{{contact}}` | Contact email address used for the response: the location's if available, otherwise the brand's. | | `{{signature}}` | Signature used for the response: the location's if it exists, otherwise the brand's. | | `{{customer_first_name}}` | Customer's first name (if available). | | `{{customer_last_name}}` | Customer's last name. | | `{{customer_full_name}}` | Customer's full name (first name + last name). | | `{{customer_civility}}` | Customer's title (e.g.: Mr., Ms., Dr., etc.). | | `{{customer_feedback_count}}` | Total number of feedbacks left by the customer. | | `{{customer_purchase_count}}` | Total number of orders placed by the customer. | | `{{customer_purchase_count_1}}` | Customer order count + 1 (useful for anticipatory wording). | | `{{customer_purchase_amount}}` | Total cumulative amount of the customer's orders. | | `{{customer_average_rating}}` | Average rating given by the customer. | | `{{purchase_amount}}` | Amount of the relevant order. | | `{{purchase_external_id}}` | External order ID. | | `{{now}}` | Lets your AI agent know the current date and time at the moment of generating a response. This can be useful if your AI agent needs to respond according to certain logic based on the current date. | ### Variables specific to satisfaction surveys Use these variables to give context to your agent responsible for generating responses for your satisfaction surveys. | Variable | Description | | ------------------------- | ------------------------------------------------------ | | `{{survey_fields}}` | List of all survey fields. | | `{{survey_csat_smileys}}` | Results of questions of type *CSAT Smileys*. | | `{{survey_nps}}` | Results of questions of type *NPS*. | | `{{survey_csat_stars}}` | Results of questions of type *CSAT Stars*. | | `{{survey_notations}}` | Set of rating-type fields (CSATs & NPS). | | `{{survey_choices}}` | Set of choice-type fields (dropdown, checkbox, radio). | ### Variables specific to contact forms Use these variables to give context to your agent responsible for generating responses for your contact forms. | Variable | Description | | ------------------ | ----------------------------------------------- | | `{{site_choices}}` | Set of choice-type fields (dropdown). | | `{{site_texts}}` | Set of text-type fields (text & purchase). | | `{{site_fields}}` | Full set of fields (dropdown, text & purchase). | ### Custom variables To further customize your AI agents, you can define [custom variables](/documentation/documentation-en/automations/variables.md) whose value will be replaced depending on the brand or location concerned. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.fullwhere.com/documentation/documentation-en/automations/ai-agents.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.