Glossary
Key terms and concepts in Ask0. Essential definitions for understanding projects, sources, knowledge bases, and AI assistant features.
Understanding the key terms and concepts in Ask0 will help you navigate the platform effectively. This glossary provides clear definitions and links to detailed documentation.
Core Concepts
Assistant
The AI-powered chat interface that answers user questions. The assistant uses your configured knowledge sources to provide accurate, contextual responses. Learn more →
Project
A container for your AI assistant configuration, including sources, settings, branding, and analytics. Each project represents a separate product or website with its own assistant. Learn more →
Organisation
The top-level account structure that contains multiple projects and team members. Organisations allow you to manage access control, billing, and shared resources across projects. Learn more →
Knowledge Management
Sources
The content and data that your AI assistant learns from to answer questions. Sources can include websites, documentation, GitHub repositories, Discord channels, files, and more. Learn more →
Web Crawler
An automated tool that discovers and indexes content from websites. The crawler follows links, respects robots.txt, and can be configured with include/exclude patterns. Learn more →
Knowledge Base
The indexed and processed collection of all your sources. The knowledge base is what the AI searches through to find relevant information for answering questions.
Embedding
The process of converting text into numerical vectors that capture semantic meaning. Embeddings enable the AI to find conceptually similar content even when exact keywords don't match.
Indexing
The process of processing, chunking, and storing content from sources in a searchable format. Indexing happens automatically when you add or update sources.
User Interface
Chat Widget
The embeddable UI component that users interact with to ask questions. The widget can be customized and positioned according to your needs. Learn more →
Commander Mode
An alternative UI mode where the chat opens as a command palette in the center of the screen, similar to Spotlight or command launchers.
Theme
The visual styling system for the chat widget, including colors, fonts, and spacing. Ask0 supports light/dark modes and custom themes. Learn more →
Integration Components
JavaScript Snippet
The code you add to your website to load and initialize the Ask0 chat widget. The snippet includes your project ID and configuration options. Learn more →
Project ID
A unique identifier for your project, used in the JavaScript snippet to connect the widget to the correct assistant and configuration.
API Key
A secure token used for server-side integrations and API access. API keys allow programmatic access to Ask0 features.
Analytics & Monitoring
Conversations
The complete chat sessions between users and your AI assistant. Conversations can be reviewed, analyzed, and used to improve responses. Learn more →
Feedback
User reactions and ratings on AI responses, including thumbs up/down and detailed feedback. This helps measure assistant effectiveness.
Engagement Metrics
Statistics about how users interact with your assistant, including number of conversations, messages per session, and resolution rates.
Source Performance
Analytics showing which knowledge sources are being used most frequently and effectively to answer questions.
Gap Detection
Analysis of questions that couldn't be answered well, helping identify missing content in your knowledge base.
Technical Terms
Vector Database
A specialized database that stores embeddings and enables fast semantic search. Ask0 uses vector search to find relevant content quickly.
RAG (Retrieval Augmented Generation)
The technique Ask0 uses to answer questions by first retrieving relevant content from your knowledge base, then using that context to generate accurate responses.
Chunking
The process of breaking large documents into smaller, semantically meaningful segments that can be indexed and retrieved effectively.
Semantic Search
Search based on meaning and context rather than just keyword matching. This allows the AI to find relevant information even with different phrasing.
Multi-tenancy
The ability to serve multiple projects and organisations from the same Ask0 instance while keeping data isolated and secure.
Deployment Options
Cloud Hosting
Using Ask0's managed service at ask0.ai, where we handle infrastructure, updates, and scaling for you.
Self-Hosting
Running Ask0 on your own infrastructure using Docker or Kubernetes. Gives you complete control over data and deployment. Learn more →
Hybrid Deployment
A combination approach where some components (like the admin console) use the cloud while sensitive data remains on-premise.
AI Models
LLM (Large Language Model)
The AI model that generates responses to user questions. Ask0 supports multiple LLMs including GPT-4, Claude, and open-source alternatives.
Embedding Model
The model that converts text into vectors for semantic search. Multi-lingual embedding models enable support for multiple languages.
BYOM (Bring Your Own Model)
Ask0's flexibility to use any AI model provider or self-hosted models instead of being locked to specific vendors.
Access Control
Members
Team members who have access to your organisation or specific projects. Members can have different roles and permissions. Learn more →
Roles
Permission levels that define what actions members can perform (e.g., Admin, Editor, Viewer).
SSO (Single Sign-On)
Enterprise authentication feature allowing team members to log in using existing corporate credentials.
Additional Terms
Webhook
HTTP callbacks that notify your systems when events occur in Ask0 (e.g., new conversation, feedback received).
Rate Limiting
Controls on API usage to ensure fair use and system stability. Different tiers have different rate limits.
Context Window
The amount of relevant information provided to the AI when generating a response. Larger context windows can provide more accurate answers.
Hallucination
When AI generates plausible-sounding but incorrect information. Ask0 minimizes this by grounding responses in your actual content.
Can't find a term? This glossary is continuously updated. If you encounter a term not listed here, please let us know or check the relevant documentation section.