Widget
Last updated
Last updated
The widget is an agent that you can integrate into your website or e-commerce, acting as an informant and a customer service available 24/7. Swiftask makes it easy to create this widget, even if you have no programming skills. Integrating the agent into your site is easy thanks to the links provided during the creation of your agent.
Configure the main "instructions" with explicit control parameters.
Implement complementary "skills" to expand the agent's functionalities.
Check the category to find the skill you need. If it's not available among the ready-to-use skills, create a new one.
Attach verified "data" if operations require documentary references.
✅ Data plays a fundamental role by providing the AI agent with a customized knowledge base. This feature, called "knowledge base," enhances the agent's capabilities by giving it access to specific information that is not part of its initial training.
Activate the "Long Term Memory" function to ensure continuity of interactions while adhering to data retention policies. This feature enables AI agents to remember past conversations with users, delivering more personalized and contextually relevant responses across multiple sessions.
This is the section where the agent can be transformed into a widget. A Swiftask widget is an AI agent that can be embedded on your website, acting as a 24/7 customer service representative, responding to visitors according to your customized instructions.
Activation and deactivation of the widget on the site
Limitation of the number of messages
Authorization for file sending
Insertion of the Chatbot name
Placement of the subtitle to display
Entry of the placeholder
Registration of the welcome message
Choice of the chatbot avatar
Selection of the chatbot interface color
Preference on the shape of the chatbot icon
Widget positioning: left or right
Brand hiding
Default opening
Welcome message
Startup questionnaires
Handling of the no-code integration code
The Widget Inbox centralizes all user conversations with your AI agents. It provides:
Chronological list of all user discussions
Conversation details (date, time, duration, message count)
Quick identification of friction points and common questions
Filtering options by date, status, keywords, and user satisfaction
Search functionality for specific conversations
This functionality is essential for several reasons:
Verifying the quality of responses: It allows you to test how your agent responds to different questions or scenarios to ensure that it provides accurate and relevant information.
Identifying weaknesses: By testing your agent with various queries, you can pinpoint areas where it lacks knowledge or responds inappropriately.
Optimizing instructions: The results of evaluations help you refine your agent’s instructions and parameters to enhance its performance.
Validation before deployment: This step ensures that your agent is ready to interact with real users and accurately represents your brand or service.
Comparing versions: You can assess different configurations of your agent to determine which one best meets your specific needs.
Even without any programming knowledge, you can easily integrate your agent into your website as a widget. To do this, once your agent is configured, go to the "Developer" section where you will find the codes to copy and paste onto your site.
Agent Client Token
Purpose: This token is a unique identifier for your agent. It is used for authentication purposes when making requests to Swiftask's backend services.
Usage: Ensure that this token is kept secure, as it grants access to your agent's functionalities.
Example: "23521f93-7487-496b-9dac-bc9db0e69cae"
Agent Slug
Purpose: The slug is a user-friendly identifier for your agent, typically used in URLs and API calls. It should be unique within the Swiftask ecosystem.
Usage: This slug can be referenced when embedding the agent in your website or when making API requests.
Example: "documentation-agentik3cc"
Widget Integration: Integration code to add the widget to your website.
Iframe Integration: Code to embed the agent in an iframe.
API Integration: It allows developers to interact with Swiftask agents programmatically using the provided client token and endpoint, enabling custom implementations that go beyond the standard widget or iframe options through RESTful API calls.
Conversation Starters : Customize the beginning of the conversation to provide a tailored welcome.
Temperature : Adjusts the bot’s creativity: high values result in varied but unpredictable responses, while low values produce more consistent responses.
Top K chunk : Defines how many parts of documents the system retrieves from the indexed ones. This helps limit the number of results and speeds up the search.