Conceptual Guide¶

This guide provides explanations of the key concepts behind the Open Chat Studio platform and AI applications more broadly.

The conceptual guide does not cover step-by-step instructions or specific examples — those are found in the How-to guides.

Terms¶

Assistant: A chatbot that uses OpenAI`s Assistant API. Assistants can write and execute code and search and reference information in uploaded files.
Authentication Provider: Authentication providers allow you to authenticate with external systems to access data or services.
Channel: The platform through which a chat occurs (e.g., WhatsApp, Telegram, Web, Slack).
Chatbot: A chatbot in Open Chat Studio is a bot configuration that interacts with users across one or more channels.
Consent Forms: Forms that provide context to chatbot users on how their data will be used and who to contact regarding any concerns.
Custom Actions: Custom actions extend a bot's functionality by integrating with external systems and APIs.
Evaluations: Evaluations is a testing system for measuring chatbot performance against different metrics.
Events: Events are a way to trigger actions in the bot based on specific conditions.
Large Language Models (LLMs): Large language models are a type of AI model that can generate human-like text, images and audio.
Messaging Provider: Messaging providers hold the configuration required to send messages to users on a specific channel.
Node: A node is a single step in a chatbot's pipeline.
Participant Data: Data that persists across sessions and is tied to the same User, Channel, Chatbot scope. It helps retain long-term user preferences and contextual information beyond a single session.
Pipelines: A pipeline is the visual workflow that controls how a chatbot processes input and produces output.
Prompt: A prompt is the instructions that are given to the LLM to generate a response. Prompts can include text, source material, and other variables.
Session: The scope of conversations between a user and a chatbot within a specific channel. Sessions are isolated, ensuring data privacy and contextual continuity for the duration of an interaction.
Session Status: The lifecycle state of a chat session (for example, setup, active, pending review, complete) and the transitions between those states.
Source Material: Additional information that can be included in the bot prompt using the {source_material} prompt variable.
Tools: Tools let your chatbot take real actions during a conversation — such as performing calculations, scheduling reminders, or storing information about a participant — rather than only responding with text.
Tracing: Tracing captures a chatbot's inputs, outputs, and decision-making process so bot developers can understand and debug their chatbot's behavior.
Versions: The ability to create and manage different versions of a chatbot.