# Overview

**GLChat** is an **enterprise-grade Agentic AI chatbot platform** built with a **privacy-first design**, offering organizations the flexibility to deploy secure, white-labeled AI assistants across web, mobile, and on-premise environments. It seamlessly integrates with existing enterprise systems, enabling a unified ecosystem where chat interfaces, administrative tools, and developer APIs work together to deliver intelligent, scalable, and branded conversational experiences.

The platform is composed of three core components that serve different user roles and technical needs:

* The **GLChat Project User Interface (UI)** — the primary interface for end users to interact with AI Agents, knowledge bases, and enterprise workflows.
* The **Admin Dashboard** — the centralized control panel for administrators to configure, monitor, and manage the GLChat environment.
* The **GLChat SDK** — a comprehensive set of APIs, language bindings, and documentation for developers to integrate GLChat capabilities directly into their projects or custom workflows.

Together, these components form a cohesive system that supports organizations throughout the full lifecycle of AI deployment—from configuration and integration to live interaction and continuous improvement.


# Overview

**GLChat** is an enterprise-grade Agentic AI chatbot platform designed with a **privacy-first** approach. It offers **easy integration** with existing software, **flexible deployment**, and **white-label** customization to provide secure, scalable, branded customer experiences. The GLChat User Interface resembles a chatbot application, allowing interaction with various features like the knowledge base, LLM, and agents.


# Introduction

The **GLChat Project User Interface (UI)** serves as the primary workspace where users interact with the AI system through an intuitive and powerful interface. Designed with a balance of simplicity and flexibility, it allows organizations to customize their chatbot experiences, manage conversation settings, and integrate with external tools, all within a consistent visual and interaction framework. Whether deployed on web, mobile, or embedded environments, the Project UI ensures a seamless experience for end-users engaging with AI Agents, knowledge bases, or workflows.

Beyond the chat interface, the Project UI is tightly integrated with the **Admin Dashboard**, enabling administrators to control configurations, permissions, and performance analytics from a centralized environment. This section provides an overview of how each component in the Project UI contributes to efficient communication, workflow automation, and enterprise-grade scalability—forming the front-facing layer of the broader **GLChat ecosystem**.


# Feature Overview

**Last update**: July 3, 2026

**GLChat** is not just an AI chatbot or AI agent platform. It is a privacy-first, white-label enterprise AI platform for building and managing Digital Employees — AI-powered colleagues with identity, persistent memory, role-based responsibilities, and the ability to take action across enterprise workflows.

Chatbots answer questions.\
AI agents execute tasks.\
Digital Employees own responsibilities.

With GLChat, enterprises can move from simple question-and-answer interactions to autonomous execution, where Digital Employees can coordinate multiple AI agents to support business processes continuously.

In short:

**Humans** manage **Digital Employees**.\
**Digital Employees** manage **AI agents**.\
**AI agents** execute **the work**.

This is the shift from using AI tools to building an enterprise digital workforce.

<figure><img src="/files/gXebXkXZNmNhyV4wiV8E" alt=""><figcaption><p>GLChat</p></figcaption></figure>

## Why GLChat — Key Differentiators

GLChat is designed for enterprises that need more than a basic chatbot. It combines privacy-first controls, enterprise customization, advanced AI orchestration, and secure deployment options to support more complex business needs.

These key differentiators represent GLChat’s advanced enterprise capabilities for organizations that require stronger governance, deeper automation, secure deployment, and industry-specific intelligence.

1. **Privacy and Security**\
   Protect enterprise data with privacy-first controls, role-based access, guardrails, and secure deployment options.
2. **White-Label Customization**\
   Fully customize GLChat with the customer’s own branding, identity, and enterprise requirements.
3. **Digital Employees**\
   Deploy Artificial Intelligence (AI)-powered digital colleagues with persistent memory, role-based responsibilities, and 24/7 operational capabilities.
4. **AI Governance\***\
   Establish governance, risk management, compliance, auditability, human-in-the-loop controls, and responsible AI operations for enterprise AI systems.
5. **GL Connectors**\
   Connect to almost 5,000 systems using Application Programming Interfaces (APIs), Function Calling, Skills, and Model Context Protocol (MCP) Servers.
6. **Knowledge Graph**\
   Improve answer accuracy by connecting entities, facts, relationships, and reasoning paths.
7. **On-Premise and Virtual Private Cloud (VPC) Deployment**\
   Deploy GLChat inside controlled enterprise infrastructure for stronger governance and compliance.
8. **Fine-Tuned Large Language Models (LLMs) and Small Language Models (SLMs)**\
   Use industry-specific and optimized models for financial services, legal, and other regulated domains.
9. **Annotated Government Documents**\
   Use annotated government documents with labels, metadata, ontology, and knowledge graphs for reliable regulatory answers.
10. **Code Interpreter and Computer Use\***\
    Execute code, analyze files, and let AI operate applications, browsers, and user interfaces.

***

## Core Features

GLChat also provides essential capabilities for everyday enterprise AI usage, allowing teams to build chatbots, connect knowledge sources, analyze files, support multiple AI models, and deliver richer multimodal interactions.

1. [**Conversational Artificial Intelligence (AI) Chatbots**](/glchat/application-ui/how-to-guide/navigation-and-chat-list/chat-interaction)\
   Provide natural conversational experiences for employees, customers, and enterprise users.
2. [**Contextual Conversations**](/glchat/application-ui/how-to-guide/navigation-and-chat-list/chat-interaction)\
   Continue conversations with retained context.
3. [**File Upload and Analysis**](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)\
   Upload and analyze documents, spreadsheets, Portable Document Format (PDF) files, and images.
4. [**Knowledge Base Integration**](/glchat/organization-admin/how-to-guides/knowledge-base)\
   Ask questions directly against connected enterprise knowledge bases.
5. [**Multimodal Capabilities**](/glchat/application-ui/how-to-guide/chat-input-and-responses/display-media-files-in-the-chat-response-area)\
   Support text and image inputs in one workflow for richer AI interactions.
6. [**Multi-Project Support**](/glchat/application-ui/how-to-guide/navigation-and-chat-list/select-project)\
   Support multiple projects under one account or workspace.
7. [**Application Programming Interface (API)**](/glchat/sdk/resources/api-references) **and** [**Software Development Kit (SDK) Support**](/glchat/sdk/overview)\
   Extend and integrate GLChat into enterprise systems.
8. [**Multi-LLM and SLM Provider Support**](/glchat/application-ui/how-to-guide/chat-configuration/switch-large-language-model-llm)\
   Support multiple model providers such as OpenAI, Anthropic, Gemini, Azure, Groq, and other supported providers.
9. [**Incognito Mode**](/glchat/application-ui/how-to-guide/navigation-and-chat-list/incognito-mode)\
   Run sensitive conversations without saving chat history.
10. [**Inline Highlighting**](/glchat/application-ui/how-to-guide/chat-input-and-responses/inline-highlight-pdf-from-reference)**,** [**Citations**](/glchat/application-ui/how-to-guide/chat-input-and-responses/answer-citation) **&** [**Sources**](/glchat/application-ui/how-to-guide/chat-input-and-responses/sources)\
    Show traceable answer references directly inside responses.

***

With ✨**over 100 enterprise-grade features**, we’ve organized them into clear categories to help you navigate. Each item is described in a short sentence for quick context, while the [User Guide](/glchat/application-ui/how-to-guide) offers in-depth instructions if you want to learn more

<figure><img src="/files/91SRAZKLhN613jKw4xhj" alt=""><figcaption></figcaption></figure>

Note: for features that currently in progress, refer to this [Features - Work in Progress](/glchat/application-ui/resources/features-work-in-progress) section

***

## Detailed Features 🔍

### Privacy & Deployment

1. ✨[**SSO (Same Sign-On)Login**](/glchat/application-ui/getting-started) **(e.g Google, GitHub, Microsoft)**
2. ✨[**Flexible Deployment**](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/flexible-deployment) (On Premise & Virtual Private Cloud)
3. [**Anonymize PII (Personal Identifiable Information) Data**](/glchat/application-ui/how-to-guide/chat-configuration/anonymize-personal-data)
4. ✨[**Role-Based Access Control (RBAC)**](/glchat/organization-admin/how-to-guides/user-management)
5. ✨[**Guardrails**](/glchat/organization-admin/how-to-guides/guardrail) **/ Content Moderation**
6. ✨[**Guardrails to Reduce Hallucination**](/glchat/organization-admin/how-to-guides/guardrail)
7. ✨[**Observability**](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/observability)
8. [Anonymous Logging](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/anonymous-logging)
9. ✨**FIPS (Federal Information Processing Standards) Compliance**
10. [Guest Mode](/glchat/application-ui/how-to-guide/navigation-and-chat-list/guest-mode)
11. [Organization Management](/glchat/organization-admin/how-to-guides/organization-management)
12. [User Management](/glchat/organization-admin/how-to-guides/user-management)
13. [Profile Settings per User](/glchat/application-ui/how-to-guide/settings/change-profile-setting)
14. Message Quota \*
15. [Incognito / Private Conversations](/glchat/application-ui/how-to-guide/navigation-and-chat-list/incognito-mode)
16. ✨[**Support Open Source Model**](/glchat/application-ui/how-to-guide/chat-configuration/switch-large-language-model-llm)
17. ✨Multitenancy
18. Captcha Implementation
19. Account Lockout

***

### White Label & Personalization

1. ✨[**White Label Project**](/glchat/sdk/resources/glchat-white-label/website-white-label)
2. [Build Customized Chatbot](/glchat/organization-admin/how-to-guides/project)
3. [Internationalization and Localization](/glchat/sdk/resources/glchat-white-label/website-white-label/internationalization-and-localization)
   1. Bahasa Indonesia Support
4. [Dark / Light Mode](/glchat/application-ui/how-to-guide/settings/change-theme)
5. Custom Persona (“My GPT” bot) \*
6. [Personalized Conversation Style](/glchat/application-ui/how-to-guide/settings/personalization-prompt-tone-and-style)
7. [Adjust Writing Style and Tone](/glchat/application-ui/how-to-guide/settings/personalization-prompt-tone-and-style)
8. [Understand Chat History ](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#understanding-chat-history)(Short-Term Memory)
9. ✨[**Long-Term Memory**](/glchat/application-ui/how-to-guide/settings/memory-management)
10. [Feedback Interaction to Memory](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#thumbs-up-thumbs-down) (Thumbs-up Thumbs-down)
11. [Personalization Prompt, Style and Tone](/glchat/application-ui/how-to-guide/settings/personalization-prompt-tone-and-style)

***

### Advanced AI Features

1. [Multi-LLM Switching](/glchat/application-ui/how-to-guide/chat-configuration/switch-large-language-model-llm)
2. [RAG with Source Highlight](/glchat/application-ui/how-to-guide/chat-input-and-responses/sources)
3. ✨[**Deep Research**](/glchat/application-ui/how-to-guide/chat-configuration/deep-research)
4. Extended Thinking \*
5. ✨[**AI Agent Integration**](/glchat/application-ui/how-to-guide/chat-configuration/select-ai-agent)
6. ✨[**AI Agent Builder**](/glchat/organization-admin/how-to-guides/ai-agent)
7. ✨[**Agent-to-Agent (A2A)**](/glchat/organization-admin/how-to-guides/ai-agent#add-remote-agent)
8. ✨[**No-Code Agent via LangFlow**](/glchat/organization-admin/how-to-guides/ai-agent#add-langflow-agent)
9. [Evaluator](https://gdplabs.gitbook.io/sdk/tutorials/evaluation/glchat-evaluation-tutorial)
10. [Document Processing Orchestrator Pipelines](/glchat/organization-admin/how-to-guides/pipeline-and-preset/dpo-pipelines)
11. ✨[**Knowledge Graph**](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/knowledge-graph)
12. [Generate Clickable Mindmap from Document](/glchat/application-ui/how-to-guide/chat-input-and-responses/generate-clickable-mindmap-from-document)
13. [Semantic Cache](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/semantic-cache)
14. [Smart FAQ Generator](/glchat/organization-admin/how-to-guides/project/manage-faq)
15. ✨[**Browser Use**](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/browser-use)
16. [Multi-Modal Conversations](/glchat/application-ui/how-to-guide/chat-input-and-responses/display-media-files-in-the-chat-response-area)
17. [Audio Input (STT)](/glchat/application-ui/how-to-guide/chat-configuration/chat-input-using-speech)
18. [Audio Output (TTS)](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#read-aloud)
19. Language Model Fine-Tuning
20. [Code Interpreter](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/code-interpreter)
21. [Search using Connectors](/glchat/application-ui/how-to-guide/settings/connector-as-search-option)
22. [Deep Research using Connectors](/glchat/application-ui/how-to-guide/chat-configuration/deep-research)
23. Self Hosted Deep Research
24. [Chat with Connectors](/glchat/application-ui/how-to-guide/chat-configuration/chat-with-connector)
25. [Smart Mode](/glchat/application-ui/how-to-guide/chat-configuration/smart-mode)
26. [Smart Hub](/glchat/application-ui/how-to-guide/chat-configuration/smart-hub)

***

### Extensibility & Integration

1. [Developer API](/glchat/sdk/resources/api-references)
2. [Software Development Kit (SDK)](/glchat/sdk/overview)
3. [Custom Tools](/glchat/organization-admin/how-to-guides/ai-agent/tools)
4. Pre-built Templates \*
5. [Embeddable Widget](/glchat/organization-admin/how-to-guides/project#step-3-deployment)
6. Headless / CLI Mode
7. [Connect to Knowledge Base](/glchat/organization-admin/how-to-guides/knowledge-base)
8. [Third-Party Connectors](/glchat/organization-admin/how-to-guides/integration-and-services/connector)
9. [WhatsApp Integration](/glchat/organization-admin/how-to-guides/project#step-3-deployment)
10. iOS
11. Android
12. Smart TV \*
13. ✨[**Model Context Protocol (MCP)**](https://gl-docs.gitbook.io/bosa/bosa-platform/bosa-connector/mcp)
14. ✨[**NLWeb**](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/nlweb)
15. [GLChat Live](/glchat/application-ui/how-to-guide/external-and-advanced-integrations/glchat-live)
16. Semantic Chunking
17. Source Code Access
18. Legacy App Integration \*

***

### Core Chat Experience

1. [Basic Conversation](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction) (Send, Edit, Copy, Regenerate Answer)
2. [Markdown Support](/glchat/application-ui/how-to-guide/chat-input-and-responses/markdown-support)
3. [Export Chat as PDF](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#export-response)
4. [Export Chat as DOCX](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#export-response)
5. [Quote & Reply](/glchat/application-ui/how-to-guide/chat-input-and-responses/quote-and-reply)
6. Pin / Star Conversation \*
7. [Notifications](/glchat/application-ui/how-to-guide/navigation-and-chat-list/notification-system)
8. [Activity & Thinking Detail](/glchat/application-ui/how-to-guide/chat-input-and-responses/show-activity-and-thinking)
9. [Step Indicator](/glchat/application-ui/how-to-guide/chat-input-and-responses/step-indicator)
10. Project / Grouping Chat \*
11. Multi-user Chat
12. [Shareable Chat](/glchat/application-ui/how-to-guide/settings/manage-shared-chat)
13. [Save & Continue Conversation from Shared Chat](/glchat/application-ui/how-to-guide/navigation-and-chat-list/chat-interaction#continue-from-shared-chat)
14. [Thumbs Up / Down Feedback](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#thumbs-up-thumbs-down)
15. Autosuggest & Autocomplete
16. [Display Media](/glchat/application-ui/how-to-guide/chat-input-and-responses/display-media-files-in-the-chat-response-area) (Image, Video, Youtube)
17. [Display Charts](/glchat/application-ui/how-to-guide/chat-input-and-responses/display-chart-in-the-chat-response-area)
18. Canvas Workspace \*
19. Help Center
20. [Switch Project](/glchat/application-ui/how-to-guide/navigation-and-chat-list/select-project)
21. [Manage Conversation](/glchat/application-ui/how-to-guide/navigation-and-chat-list/chat-interaction) (Search, Rename, Archive)
22. [Task Scheduler](/glchat/application-ui/how-to-guide/chat-configuration/task-scheduler)
23. [Question Library](/glchat/application-ui/how-to-guide/chat-input-and-responses/question-library)
24. [Slash and Mention Command](/glchat/application-ui/how-to-guide/chat-input-and-responses/slash-and-mention-command)
25. [Start as Separate Chat (branch message to new conversation)](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#start-a-separate-chat-branch-message-to-a-new-conversation)
26. [AI Agent as Default Interaction Mode](/glchat/application-ui/how-to-guide/chat-input-and-responses/ai-agent-as-default)
27. [AI Agent with User Authentication](/glchat/application-ui/how-to-guide/chat-input-and-responses/agent-with-user-authentication)
28. Mermaid Diagram Display
29. Guest Mode Project
30. [HTML Preview in GLChat](/glchat/application-ui/how-to-guide/chat-input-and-responses/html-preview-in-glchat)
31. [Interactive Agent Responses](/glchat/application-ui/how-to-guide/chat-input-and-responses/interactive-agent-responses)

***

### File & Multimodal Support

1. [Document (DOCX, PDF) Upload & Preview](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
2. [Spreadsheet files (CSV, Excel)](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
3. [Presentation file (PPTX)](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
4. [Upload Multiple File Types](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
5. [Image Upload & Analysis](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
6. [Audio Upload & Analysis](/glchat/application-ui/how-to-guide/chat-configuration/upload-document-file)
7. Youtube URL Analysis
8. Scrap Website URL
9. OCR (Image to Text) \*
10. Table & Diagram Reader \*
11. [Highlight Used Text in PDF](/glchat/application-ui/how-to-guide/chat-input-and-responses/inline-highlight-pdf-from-reference)
12. [Rich Code Display](/glchat/application-ui/how-to-guide/chat-input-and-responses/markdown-support)
13. [Web Search](/glchat/application-ui/how-to-guide/chat-configuration/web-search)

\
(\*) Feature will be coming soon


# Getting Started

This is the official guide to using **this conversational platform** (powered by **GLChat**, developed by **GDP Labs**).

This platform is designed to help you interact with information and services more efficiently through natural conversation. Whether you're a first-time user or someone looking to explore advanced features, this guide will walk you through everything you need to know to get the most out of it.

Let’s get started!!

Navigate to the platform link. For the authentication process, the platform only allows user accounts that are already set before to access this project. The platform also provides login via single sign on (depends on your setup). Currently supporting:

1. Basic authentication (username and password)
2. Sign in with Google
3. Sign in with Azure Entra ID
4. Sign in with GitHub

<figure><img src="/files/UIY2QLzSJXWPWgPsnPuo" alt=""><figcaption></figcaption></figure>

Once you have successfully logged in, you will be redirected to the platform's homepage.

<figure><img src="/files/alGB8B5UygLrhL8lryrV" alt=""><figcaption></figcaption></figure>


# How-To Guide

**Last update**: May 13, 2026<br>

Welcome to the **GLChat User Guide**.\
This guide provides a step-by-step walkthrough of GLChat’s capabilities so you can get the most out of the platform.

While the [Feature Overview](broken://pages/P42Qfb5mQQJsxEQe4kcD) page gives you a quick snapshot of what it can do, this guide focuses on **how to use each feature in practice**. You’ll find clear instructions, screenshots, and tips for both everyday tasks and advanced use cases.

We’ve grouped features into categories for better navigation:

## Navigation and Chat List

<table data-full-width="false"><thead><tr><th width="44.8011474609375">No</th><th width="200.08453369140625">Feature</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a href="/pages/gsezCt7cAEPu8keiETTY">Select Project</a></td><td>The Project selector in the top left corner allows users to switch between Applications assigned to them in the environment</td></tr><tr><td>2</td><td><a href="/pages/o7zq5YnHYOEoidZaZrIi">Chat Interaction</a></td><td>Basic chat features, such as creating a new chat, renaming a chat, sharing a chat, and deleting a chat</td></tr><tr><td>3</td><td><a href="/pages/K9f83YXcS977yqUg984X">Search Chat</a></td><td>Search chat or conversation titles across assigned projects based on the term entered in the search box</td></tr><tr><td>4</td><td><a href="/pages/WBZkqUuERxKYTMkc2YC2">Delete All Chats</a></td><td>Capability to delete all conversations in the project</td></tr><tr><td>5</td><td><a href="/pages/ER1vXDT34qg8OvEiEGf8">Incognito Mode</a></td><td>A mode that allows users to ask questions without the system recording the history of the chat during incognito mode</td></tr><tr><td>6</td><td><a href="/pages/KZrKlLdQgXziCPRoODw9">Notification System</a></td><td>Provide users with updates on completed tasks and suggest the next actionable steps</td></tr></tbody></table>

## Chat Configuration

<table><thead><tr><th width="44.9964599609375">No</th><th width="199.9722900390625">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a href="/pages/oTdTNFT5Z6BWQZs5WERU">Select AI Agent</a></td><td>Integration with AI Agents for additional capabilities depending on the agent's features</td></tr><tr><td>2</td><td><a href="/pages/zLRZdaRg7pFz9scPZRNY">Upload Document / File</a></td><td>Capability to upload documents or files supported by the system for questioning in the project or as additional references</td></tr><tr><td>3</td><td><a href="/pages/AwMTjKFy5vJiJKDphCyy">Web Search</a></td><td>Capability to retrieve information or context from the internet as knowledge to answer questions when triggered</td></tr><tr><td>4</td><td><a href="/pages/8xDBOCRGccomqx00UkD9">Deep Research</a></td><td>Unlock insights with our deep research feature, diving into comprehensive data analysis and uncovering hidden patterns to inform your decisions.</td></tr><tr><td>5</td><td><a href="/pages/KTqVvA6RW2RdzCQOaAkC">Switch Large Language Model (LLM)</a></td><td>The capability to switch between different LLMs supported in the project</td></tr><tr><td>6</td><td><a href="/pages/NWXk6Osz6A0zTKfqIuTa">Chat Input Using Speech</a></td><td>Capability to chat using Speech-to-Text technology that interprets speech into messages in the chat input</td></tr><tr><td>7</td><td><a href="/pages/DC0LC60DuNboUdOqVMWk">Recommended Question per Project</a></td><td>Capability to set recommended questions for each project, providing users ideas on what questions can be asked of a specific application</td></tr><tr><td>8</td><td><a href="/pages/Q7bgvXPQyBQwcAnD3HSA">Anonymize Personal Data</a></td><td>A toggle in the Chat Control (right sidebar) to determine whether users want to anonymize or deanonymize PII in the chat before sending it to the LLM</td></tr><tr><td>9</td><td><a href="/pages/wunhLojgFXDzwq87mypK">Speech Configuration</a></td><td>Configure the language used for speech input and read-aloud features to enable more natural voice interactions.</td></tr><tr><td>10</td><td><a href="/pages/lRKjIzHa9SRdbCLrQX8Y">Chat Filter</a></td><td>The capability to filter chat messages is based on configurations set in the DPO and a reference CSV file. This allows for customized filtering to meet specific requirements.</td></tr><tr><td>11</td><td><a href="/pages/4PTpLfQqI04PcKsBYLE2">Schedule Task</a></td><td>Capability to schedule a task and execute it based on a preferred schedule.</td></tr></tbody></table>

## Chat Input and Responses

<table><thead><tr><th width="45.24365234375"></th><th width="199.96661376953125">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a href="/pages/sXrSEKEcln6A24cYD5b9">Step Indicator</a></td><td>Shows the steps and processes that occur from the moment the user sends a message until a response is generated. It will show the steps involved from backend and also the total duration</td></tr><tr><td>2</td><td><a href="/pages/8ASqGhGECiIMZc8Ncefh">Message Interaction</a></td><td>Basic message interaction such as starting a new message, editing a message, copy response, regenerate response, and providing feedback on the response (thumbs up and thumbs down)</td></tr><tr><td>3</td><td><a href="/pages/2tydET5J6dFHJuOVHQTT">Quote and Reply</a></td><td>Feature that allows users to quote specific answers from the bot and use them as context for future chats</td></tr><tr><td>4</td><td><a href="/pages/PeikJ5nqVUh0i6ce93WQ">Display Media Files in the Chat Response Area</a></td><td>Capability to display media files (currently images and YouTube widgets) inline in the chat area based on contextual answers</td></tr><tr><td>5</td><td><a href="/pages/SHHLVKEv9yzrGnYhDSo9">Display Chart in the Chat Response Area</a></td><td>Capability to display chart inline in the chat area based on contextual answers</td></tr><tr><td>6</td><td><a href="/pages/nYB7V6hLX2VcRr2cjaYV">Inline Highlight PDF from Reference</a></td><td>Capability to preview PDF files inline instead of opening them in a new tab, and to highlight the relevant sections used as context during the answer to the question</td></tr><tr><td>7</td><td><a href="/pages/CX7GyJB2d7WWYas7Jsjo">Answer Citation</a></td><td>Feature that adds citations to the answers generated by the LLM and links or highlights the source of reference</td></tr><tr><td>8</td><td><a href="/pages/WPcFciu3Y8EIGRcUKERT">Sources</a></td><td>Capability to display relevant reference sources for answers provided by the bot</td></tr><tr><td>9</td><td><a href="/pages/6WjEU3hTzrTmsPCckN40">Related Queries</a></td><td>Feature that enables auto-suggestions of prompts to be entered, working in the New Chat area</td></tr><tr><td>10</td><td><a href="/pages/ujyZ92YsDOaTF7Fp6S83">Generate Clickable Mindmap from Document</a></td><td>Visualize document as interactive mind maps</td></tr><tr><td>11</td><td><a href="/pages/xapTXVQsUFdl3pZZRdPi">Question Library</a></td><td>Centralized list of recommended and auto-generated questions to guide conversation starters in the project.</td></tr><tr><td>12</td><td><a href="/pages/oKMGBfu61WAg4h45D9JR">Slash and Mention Command</a></td><td>Enable users to perform quick actions, invoke AI Agents, Search Deep Research, or control chat sessions directly from the chat input using intuitive /commands and @mentions</td></tr></tbody></table>

## Settings

<table><thead><tr><th width="45.17254638671875">No</th><th width="200.39630126953125">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a href="/pages/728W0r7I1OS2lvb1WajY">Change Theme</a></td><td>Ability to adjust the application theme (Dark/Light mode) via profile settings</td></tr><tr><td>2</td><td><a href="/pages/MTfS62S788KbhFBTWbQb">Change Language</a></td><td>Ability to adjust internationalization and localization preferences via profile settings</td></tr><tr><td>3</td><td><a href="/pages/HBVpQUIjhfSTl0rkKHKe">Change Profile Setting</a></td><td>Manage and update your personal profile information to keep your account details accurate and personalized.</td></tr><tr><td>4</td><td><a href="/pages/8k7pxhrbeCmYGpfAJvk5">Preference on LLM Model, Anonymize Personal Data and Default Application</a></td><td>Capability that allows users to set preferences for the LLM to be used in the chat area, as well as configure the anonymization of personal data and default application</td></tr><tr><td>5</td><td><a href="/pages/oFR8J5I4fx4kBWne6drg">Connector as Search Option</a></td><td>Search directly within your own tools and data sources (e.g., Gmail, Google Calendar, Google Drive, GitHub) instead of only relying on the web or knowledge base</td></tr><tr><td>6</td><td><a href="/pages/M7XXpaCAkbPgkzXdazAJ">Managed Shared Chat</a></td><td>Manage and organize your shared chat in one place</td></tr><tr><td>7</td><td><a href="/pages/vDfA1vYWQPo257k0tWWg">Help Center</a></td><td>Access user guides and submit reports to quickly resolve issues and ensure smooth application usage.</td></tr><tr><td>8</td><td><a href="/pages/CdXRZvFNJuVbkylnPdtj">Personalization Prompt, Tone and Style</a></td><td>Allow users to customize app responses with their chosen tone, style, and detail level to match their communication preferences.</td></tr></tbody></table>


# Navigation and Chat List


# Select Project

In GLChat, a project functions as a repository for chatbot configurations, encompassing the knowledge base, AI agent, and other components or features.

<figure><img src="/files/zVjTuJS9LLxNYmCaJQ7y" alt=""><figcaption></figcaption></figure>

If you are **assigned to more than one project**, an project selector will be available for you to switch between them. You can search the project using the keyword, or based on AI Agent that is assigned to the project

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/d7lUiqYK2SydpPGZXh7V" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/Wadjgy2eTMSi4WVKE0q0)
{% endtab %}
{% endtabs %}

For example, as shown in the image, users can select between "Demo General Purpose", "Demo Media Mapping", or "Regulatory Compliance with RAG" from the selector.

Note that if you only have one project assigned, you will not be able to change project and the UI would look like this.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/CDlLKCUpJGcfDcshdy9e" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %} <img src="/files/29Fp1w9Sj9xvVMzgI14x" alt="" data-size="original">
{% endtab %}
{% endtabs %}


# Chat Interaction

## New Chat

After you have logged in and choose your project, you can start a new conversation by entering your message in the text box. You can customize your chat experience through the [chat configurations](/glchat/application-ui/how-to-guide/chat-configuration).

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/aDReaHOCQ6DCqPe8aKwp" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %} <img src="/files/ZpntPsAtVzbFW5it328I" alt="" width="375">
{% endtab %}
{% endtabs %}

Click the Send button or pressing Enter on your keyboard to submit your message to start the conversation.

Or, you can easily start a new conversation by clicking the “New Chat” button on the top-left sidebar.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/quC0nCkxCpOXZMgI4ZFN" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/grFwVYQdZVIE522pEd4t)
{% endtab %}
{% endtabs %}

## Auto Rename New Chat

After submitting your message, the project will process your message and your conversation will be automatically renamed according to your first message.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/e7YyACtMd1djcno6GYav" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/5okedwK5HmnPix7nMuJb)
{% endtab %}
{% endtabs %}

## Rename Chat

You can also manually **rename the chat/conversation title** by clicking the **three dots** menu on the selected conversation and choosing the **Rename** option.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/UezsmARjBe0AbLX2AFhz" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/CXEiQCRZWC85TtJvZi3t)
{% endtab %}
{% endtabs %}

A dialog box will appear; simply type your new title and click the **Rename** button.

**Note: The new title cannot be left blank.**

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/EGzDsDLBetEezsCObNIN" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/8GJioQEv2dmo0wEYtexl)
{% endtab %}
{% endtabs %}

## Share Chat

You can share your conversation by **clicking the three dots** menu on the selected conversation and choosing the **Share** option.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/Ze1P5DqpXa50rjtUvVIR" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/EPsvrI1Q5xZT9CIQlXzm)
{% endtab %}
{% endtabs %}

It will prompt confirmation modal showing the link of the conversation that you wish to share. Click **Copy Link** button and share the conversation link.

**Note: The shared conversation is a snapshot taken when you first share it or when you click the `Update link` button. Later messages on that conversation remain private, unless you update the link.** See [Update Shared Chat](#update-shared-chat) section.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/SrnfJBzsTivG8PgylJLE" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/tUrtjwzBWllQbRsClp3h)
{% endtab %}
{% endtabs %}

If the recipient is **not logged in** (or does not have an account), they will access the chat in **guest mode**. In this mode, the conversation is **read-only** and cannot be continued.

If the recipient is a **registered user and logged in**, they will be able to view the shared conversation normally and can [continue from the shared chat](#continue-from-shared-chat).

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/QO9qGGY2D8alBKTIM3j7" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/2dFBJx0Np9aZSi77o3LP)
{% endtab %}
{% endtabs %}

## Update Shared Chat

If you already share a chat before and you want to update the shared chat, you can simply go to Share button and click Update link. The link share chat will be updated with latest content.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/SrnfJBzsTivG8PgylJLE" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/Vur3grpUl5x8OCgDJJkD)
{% endtab %}
{% endtabs %}

## Continue from Shared Chat

If you are on a shared chat, whether it was yours or from someone else's, now you can continue the chat and it will be **created as a new chat** under your account with the same title as the Shared Chat. Simply type and send prompt from a shared chat and it will be created.

<figure><img src="/files/p9REuLLKo2MPAKs9payN" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/p9REuLLKo2MPAKs9payN" alt=""><figcaption></figcaption></figure>

## Delete Chat

You can also delete a chat/conversation by **clicking the three dots** menu on the selected conversation and choosing the **Delete** option.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/gBPpocrtY9g1a9BU3K6e" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/sf2HHuXzeU6AW0p0qgEE)
{% endtab %}
{% endtabs %}

A dialog box will appear; simply click the **Delete** button.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/ZCwn7tFr5m9AeyM0sks1" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/WYM9Y2b4NVrLRqwaMSUV)
{% endtab %}
{% endtabs %}

If you delete a conversation, any previously shared link for that chat will no longer display the conversation content. Recipients who open the old link will see an empty page with a message that the conversation couldn't be found.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/m0KZr2ChFFozv9Yflkg3" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/YXpRynrcgsfJpsHPk2Bp)
{% endtab %}
{% endtabs %}


# Search Chat

You can search your previous conversations or chats by clicking the **Search** button on the top-left sidebar and typing a keyword.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/9p0xCL7EBw1Iy6Q8rBxN" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/J9rVBgoDS97LKANrgG0A" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/uyQEpPtEHTxKXNhiVsHB) ![](/files/cVkO2Kb7nUL0q4IOVizQ)
{% endtab %}
{% endtabs %}

It will display all conversations that contain the keyword, either in the conversation title or within the chat content.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/BDjPHbg72o38rLAl6vRb" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/X1XLxwlJ9HzuevzCQ5fy)
{% endtab %}
{% endtabs %}

If you click the chat, you will be directed to that conversation.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/SxwISeidRwskAdH3sjxl" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/aAFJDPDcYZadBTGXjlKb)
{% endtab %}
{% endtabs %}


# Delete All Chats

> Note: This feature is not available on mobile devices.

You can delete all conversations by clicking **View All** from the left sidebar and select the conversation that will be removed by click **Select**.

<figure><img src="/files/8YIrIFdmvgKx5DwiuDgY" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/22EZ4sUjHpxR0aKyTXfj" alt=""><figcaption></figcaption></figure>

Once you click Select, you can choose whether you want to delete specific conversation or you want to **Select All.**

<figure><img src="/files/PBAMuJ942SmlywiMaWoq" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/oDEVieYpILwugfbA1mz2" alt=""><figcaption></figcaption></figure>

After choosing the chats you want to delete, click the Delete button, then a confirmation pop up will appear. If you're already sure, click Delete again.

<figure><img src="/files/pZClxnJ7Rn2i6fvPKLIY" alt=""><figcaption></figcaption></figure>

This will remove all chat history.


# Incognito Mode

Incognito Mode is designed for temporary conversations. It allows you to use GLChat without saving the conversation to chat history or showing it in the chat list after the session ends.

When you use Incognito Mode:

* Your conversation is **not saved to chat history.**
* Your conversation will not appear in the chat list after you leave or end the session.
* You cannot continue, reopen, or share the same incognito conversation later.
* Previous incognito conversations are not reused as future conversation context or memory.
* You can still ask questions and use supported GLChat capabilities, depending on the selected Project configuration.

However, Incognito Mode is not the same as full anonymous browsing, automatic personal data masking, or automatic prevention of sensitive data input.

### Summary of Normal vs Incognito Mode

| Feature                                         | Normal Mode                                     | Incognito Mode                                  |
| ----------------------------------------------- | ----------------------------------------------- | ----------------------------------------------- |
| Ask questions and receive answers               | ✅ Available                                     | ✅ Available                                     |
| Knowledge Base answers                          | ✅ Available if the Project has a Knowledge Base | ✅ Available if the Project has a Knowledge Base |
| Search, Deep Research, Connectors, and AI Agent | ✅ Available depending on Project settings       | ✅ Available depending on Project settings       |
| Chat history                                    | ✅ Saved in chat history                         | ❌ Not saved in chat history                     |
| Chat list                                       | ✅ Appears in the chat list                      | ❌ Does not appear after the session ends        |
| Continue previous conversation                  | ✅ Can be reopened and continued                 | ❌ Cannot be reopened or continued later         |
| Share Chat                                      | ✅ Available                                     | ❌ Not available                                 |
| File upload                                     | ✅ Available depending on Project settings       | ❌ Not supported                                 |
| Thumbs Up / Thumbs Down                         | ✅ Available                                     | ❌ Not available                                 |
| Memory / future context                         | ✅ May be used depending on configuration        | ❌ Not used as future conversation context       |
| Full anonymous browsing                         | ❌ Not guaranteed                                | ❌ Not guaranteed                                |
| Automatic personal data masking                 | ❌ Not provided by default                       | ❌ Not provided by default                       |

You can enable **Incognito Mode** by clicking **Incognito** icon on the top right side during New Chat.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/86Ng6fUYsJ1DSiOAsAB5" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/tq0sCr58JcFspGAgbPNR" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<div><img src="/files/sA1dBcPUBCvDBManzuYy" alt="" width="375"> <img src="/files/ZpntPsAtVzbFW5it328I" alt="" width="375"></div>
{% endtab %}
{% endtabs %}

<br>


# Guest Mode

Guest Mode allows users to access project without logging in, enabling quick, temporary interactions while maintaining data privacy and session isolation

Guest Mode is designed for users who need instant access without requiring an account or authentication. It’s ideal for demos, trials, or public deployments where ease of access is critical. In this mode, user sessions are temporary. Once the browser is closed or the session expires, all conversation history and uploaded data are automatically cleared to protect privacy.\
\
Currently only limited features are available in Guest Mode other than conversational interaction such as: anonymize personal data and question library

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/qSPpSQPSdzXr0AtZwZWY" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/TmAD0FH26oBhKbEYUx9X" alt="" width="375"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

<br>


# Notification System

#### Notification System Overview

This project includes a notification system to inform users when a task is executed or completed. Tasks can vary, for example, an inquiry to an LLM, interaction with an agent, or deep research activities. Notifications provide users with updates on completed tasks and suggest the next actionable steps. There are 2 type of notification that currently supported

1. In App Notification
2. Email Notification

**How to Access Notifications**

* Click the 🔔 (bell) icon located in the top left sidebar and it will show several notifications (if any).

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/JFd3OarnyiDCjYMkgLgR" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/oS4Rm9SBH4ryRpvADuxG)
{% endtab %}
{% endtabs %}

* A red dot will appear on the icon if there is a new notification.
* To view more notifications, click "See More" to be redirected to the "View All Notifications" page.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/WK4VuIPLdu5HSDmoaSQk" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/6AHdsKK1T0anxUazdA7t)
{% endtab %}
{% endtabs %}

* Clicking a notification related to task scheduler will redirect user to a specific conversation

**Current Implementation**

* Notifications are currently implemented for the Task Scheduler.
* Future updates will integrate notifications directly with other components, gradually bypassing the Task Scheduler.
* Notifications will remain visible for 7 days.

#### Email Notification

Notifications are available both via in-app alerts and through email.

<figure><img src="/files/9XxilrdFP5YP6NUXDdUb" alt=""><figcaption></figcaption></figure>


# Chat Configuration


# Select AI Agent

You can select an AI Agent at the top of the chat box. If no agents are displayed, it means the project you are using has not been assigned any chatbot.

AI Agents are only available in **Chatbot Mode** (not in other modes like Web Search or Deep Research).

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/Mlc4mrRBMKxpylXvsg8I" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/1OGuS8vFROaAst2cvKs7)
{% endtab %}
{% endtabs %}

Simply choose one from the list, or click the **More Agents** button to see the complete list of available agents. You can also search for a specific agent by typing its name.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/EelZdP2uFB49b5gyVF3E" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/EelZdP2uFB49b5gyVF3E" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/s3xivEwM6ZmcDEryjlHs) ![](/files/k85Kcss0A6qAQ1iq040h)
{% endtab %}
{% endtabs %}

Each agent has specific tasks and interacts with specific tools — you can hover over an agent’s name to read its description before selecting it.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/wPzvBq4hMTJiEKWNZLRp" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/k85Kcss0A6qAQ1iq040h)
{% endtab %}
{% endtabs %}


# Upload Document / File

There are two ways of uploading a document or file:

1. **Directly in the chat box**
2. **From the Chat Control panel** on the top-right corner

### Attach File in the Chat Box

You can upload a document or file by clicking the **+** button and choose **Attach File** icon in the chat box.

<figure><img src="/files/FP1ub9imYT8Re9rAC9Dr" alt=""><figcaption></figcaption></figure>

{% tabs %}
{% tab title="Website" %}
There are two ways to add files:

1. Upload from your computer
2. Add directly from Google Drive (if your account is connected)

<figure><img src="/files/KvYoeVGeZ92ceDeZ0J6H" alt=""><figcaption></figcaption></figure>

### Upload From Computer

When you choose to upload from your computer, a window will appear where you can drag and drop your files or click **Select** to pick them manually.

You can upload up to five files at once, and each file can be up to 10 MB in size.

<figure><img src="/files/TkzN5GPbBTMW4MjPEnpM" alt=""><figcaption></figcaption></figure>

The types of files you can upload depend on both the configuration of the **Document Processing Orchestrator (DPO)** module and the **LLM you have selected**. Some examples include:

1. Image (.png, .jpg, .jpeg, .webp)
2. Spreadsheet files (.xlsx, .csv)
3. Audio files (.wav, .mp3)
4. Document files (.pdf and .docx)
5. Log or markup files (.txt, .md, and .log)
6. Source Code (.java, .py)
7. Video (supported by specific LLM such as Google's **Gemini**)

The maximum file size allowed is 10 MB per file.

### Upload From Google Drive

If you choose to add files from Google Drive, the system will prompt you to sign in and grant access the first time you use this option. After authentication, or if you have already connected your Drive before, you can immediately browse your Google Drive files and select the ones you want to attach.
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
On mobile devices:

* Tap **Files** to select and upload documents or other files from your device.
* Tap **Camera** to take a new photo and upload it.
* Tap **Photos** to choose an existing image from your gallery.

<figure><img src="/files/Hq3xcrYDo3DDbViqxFgl" alt="" width="375"><figcaption></figcaption></figure>

The types of files you can upload depend on both the configuration of the **Document Processing Orchestrator (DPO)** module and the **LLM you have selected**. Some examples include:

1. Image (.png, .jpg, .jpeg, .webp)
2. Spreadsheet files (.xlsx, .csv)
3. Audio files (.wav, .mp3)
4. Document files (.pdf and .docx)
5. Log or markup files (.txt, .md, and .log)
6. Source Code (.java, .py)
7. Video (supported by specific LLM such as Google's **Gemini**)

The maximum file size allowed is 10 MB per file.
{% endtab %}
{% endtabs %}

## Upload File from Chat Control

> Note: This feature is not available on mobile devices.

You can also attach files through the **Chat Control** panel by clicking the **Files** option in the top-right corner.

<figure><img src="/files/ZGvUHmVE3Yu48v8aBNvW" alt=""><figcaption></figcaption></figure>

This opens the file management sidebar, where you can:

* View all files you have already uploaded
* Click **Upload File** to add new files from your computer or Google Drive (the same interface as the chat box attachment)

<figure><img src="/files/Vo72RudE5kahlRTAMNRQ" alt=""><figcaption></figcaption></figure>

This method is useful if you want to upload files **before** writing your message or manage previously uploaded files in one place.

## Preview Uploaded File

Once you send a message with an attachment, you can **preview the file directly in the chat area** by clicking on the file preview.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/oiKszi2OLzCGgtAKM3JC" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/IfRGXPTicrNkMiuBQjEP)
{% endtab %}
{% endtabs %}

After that a popup will open showing the full file content.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/L6VzqkNfhJIYhLM1Uxy8" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/m3KJeBUJKPsIlDPNm8tF)
{% endtab %}
{% endtabs %}


# Web Search

You can browse the internet for more up-to-date information by enabling the **Web Search** tool.

To do this, click the **+** button, choose **Tools** in the chat box, select **Search**, and then toggle **Web** on.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/t5kwTFhszY9imXFnvD2l" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/MWAuS0507wlzW2gCMlHV) ![](/files/U9cLOottqxYFdclU4Nw3)
{% endtab %}
{% endtabs %}

Once it is enabled, simply type your question in the chat box and send it — the answer will include information from the web.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/6GUaYnd6pN3XQZSQz7Ut" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/1ppbiqGklohjyUa4W0Rh)
{% endtab %}
{% endtabs %}


# Deep Research

Unlock insights with our deep research feature, diving into comprehensive data analysis and uncovering hidden patterns to inform your decisions.

Deep Research allows you to get a more thorough, multi-step exploration of a topic compared to a standard chat response.

To enable it, click the **+** button and choose **Tools** button in the chat box, select **Deep Research**, and then choose between the available mode.

### Essentials Mode

A quicker research mode designed for high-volume use cases. It gathers key information and summarizes main points in approximately 2–10 minutes.

Best for:

* Quick decision-making
* Budget-conscious tasks
* Lightweight reasoning tasks

### Comprehensive Mode

The most thorough research mode that delivers professional-grade, in-depth results. It covers all relevant angles and details, typically running for 10–30+ minutes.

Best for:

* Multi-step research
* Comparative analysis
* Academic work
* Complex queries requiring maximum depth

### Private Mode

Private Mode is designed for organizations that require stricter data privacy and security.\
This mode uses self-hosted language models, ensuring data is processed within the client environment and is not exposed to third-party AI providers.

Best for:

* On-premise deployments
* Sensitive enterprise data
* Compliance and internal security requirements
* Finding information in company files, searching Google Drive, and organizational knowledge management

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/FlWjvcKJowsu9DkpYlEP" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<div><img src="/files/y9DaLj1syfnQoUDt3tlL" alt="" width="375"> <img src="/files/KWUq9Zn8RxbUs3bypJ5L" alt="" width="375"></div>
{% endtab %}
{% endtabs %}

Once selected, simply type your question in the chat box and send it — the system will perform a structured research process and return a richer, well-organized response.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/OqFh51xIYqsAFrs2U3gB" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %} <img src="/files/oNP5m7NqrQTHiHR0aWmp" alt="" height="415" width="192">
{% endtab %}
{% endtabs %}

***

### External and Internal Data Support

Previously, Deep Research only supported external web research.\
Deep Research now supports both:

* External Data (Web)
* Internal Enterprise Data via Connectors

This allows combining public information with organization-specific knowledge for more accurate and contextual responses.

#### Supported Internal Data Sources

Deep Research can now connect to:

* Project Knowledge Base
* Google Calendar
* Google Drive
* GitHub
* Gmail

This enables use cases such as:

* Research enriched with internal documentation
* Context-aware answers using project knowledge
* Cross-referencing emails, files, calendars, and repositories
* Internal enterprise analysis combined with external web insights

***

Current capability of Deep Research:

* **Information Retrieval:** Performs deep scans of both the internet and internal organizational data (Google Drive).
* **Multi-Step Exploration:** Conducts thorough research of the topic based on the specific prompt sent.
* **In-line Citations:** Includes clickable citations to indicate the origin of specific facts or claims, whether they come from a website or an internal document.
* **Reference Lists:** Provides a complete list of clickable references at the end of the response for easy access to original sources.


# Switch Large Language Model (LLM)

You can switch models by clicking the **Models** options in the chat box. From there, you can select the desired model.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/Vs18ZC5QuyqgIvQpjzb5" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/B4ceqm8F8J5dqUn8nUg0)
{% endtab %}
{% endtabs %}

Or, you can change the model during the message regeneration.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/ON3us99vt6AracTMic5n" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/m1K3RLZUQ3qI7YYaGgwZ)

![](/files/vM4qVWrBfEbT1uBah6QO)

![](/files/WbWFcaTR77Q0ffJWbrpZ)<br>
{% endtab %}
{% endtabs %}

### Supported Models

The following is a non-exhaustive list of available models.

1. GPT 5 Mini
2. GPT 5.1
3. GPT 5.2
4. GPT 5.4
5. GPT 5.5
6. Claude Sonnet 4.5
7. Claude Sonnet 4.6
8. Claude Opus 4.5
9. Claude Opus 4.6
10. Claude Opus 4.7
11. Claude Opus 4.8
12. Claude Haiku 4.5
13. Gemini 2.0 Flash
14. Gemini 2.5 Flash
15. Gemini 3.0 Flash
16. Gemini 3.5 Flash
17. Gemini 2.5 Pro&#x20;
18. Gemini 3 Pro
19. Gemini 3.1 Pro
20. DeepSeek V3 (671B)
21. DeepSeek R1 (671B)
22. DeepSeek V4 Flash
23. DeepSeek V4 Pro
24. Alibaba Qwen2.5 14B Instruct
25. Llama-SEA-LION-v3.5-8B
26. Llama-SEA-LION-v3.5-70B
27. Qwen 3-30B
28. Qwen 3 Next 80B - A3B
29. Qwen 3 235B - A22B - Instruct
30. Meta Llama 4 Scout
31. Meta Llama 4 Maveric
32. Grok 4 Fast
33. Grok 4.1 Fast
34. Kimi K2.5
35. Kimi K2.6
36. GLM 5
37. Gemma 4-26B-A4B IT
38. Gemma 4-31B IT

\
\*Other models are available by request.


# Chat Input Using Speech

You can also input chat using speech by clicking the **Voice Mode** icon in the chat box. Make sure to allow microphone permission. Simply speak during the recording.

**Limitation**: It only supports up to **100 characters (including spaces)** to prevent exceeding the quota.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/ljwPUmoyJvMyVIq6po9x" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/GDbcadTRbMky6DEJMltn)
{% endtab %}
{% endtabs %}


# Recommended Question per Project

This feature enables you to have pre-defined recommended questions that can be asked to a specific project. Currently, the list of questions will show a maximum of 4 in a round-robin fashion from all questions determined in the Admin Dashboard.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/1MgD74uHR4srK1gtxuRt" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/QhZqc47w34E5i7mKfFsT)
{% endtab %}
{% endtabs %}


# Anonymize Personal Data

To enable Anonymize Personal Data feature, you can adjust the settings by clicking the **+ Button** and choose **Protect Data** .

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/Sn1dmR4G35IX8e0UQUIh" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %} <img src="/files/2PIz6FEAXgZecIwfKK2J" alt="" width="375">

<br>
{% endtab %}
{% endtabs %}

Next, use the toggle switch to turn the specific options on or off as needed. The corresponding information will then be displayed to confirm the changes.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/E4SaBowsyYO4bQv9m6jJ" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %} <img src="/files/UK2pZ5e9qpr8iTFQyMgj" alt="" width="375">
{% endtab %}
{% endtabs %}

You can view anonymized data related to PII by clicking the **Mask Personal Data** icon and deanonymize the data by clicking the **Show Personal Data** icon, located at the bottom of both chat messages and responses.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/PxOhzfiaqa1zq3FOKhO5" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/Lk3iEmO8p2G3vPh9hNqd" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<div><img src="/files/S9tF9DWySNsKw8KUfDki" alt="" width="375"> <img src="/files/kMPIYwUwQIAKpC7knveB" alt="" width="375"></div>
{% endtab %}
{% endtabs %}

Currently, the supported PII types include:

1. KTP
2. NPWP
3. PROJECT
4. ORGANIZATION
5. EMPLOYEE\_ID
6. FAMILY\_CARD\_NUMBER (KK)
7. FACEBOOK\_ACCOUNT
8. LINKEDIN\_ACCOUNT
9. BANK\_ACCOUNT
10. BPJS ACCOUNT (BPJS Kesehatan)
11. EMAIL\_ADDRESS
12. PHONE\_NUMBER
13. PERSON

**Please note** that currently, we only **anonymize and deanonymize data during the generation of responses**.


# Speech Configuration

{% tabs %}
{% tab title="Website" %}
To set up or change your speech settings, click the **Chat Control** icon located in the top-right corner of the screen. From the panel that appears, select **Speech**.

<figure><img src="/files/vA887oP9DfG48I3yaFXZ" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
To set up or change your speech settings, click the **three-dot (⋯)** icon located in the top-right corner of the screen. From the panel that appears, select **Speech**.

![](/files/YoPOV5nvhp6MRMT56kzm) ![](/files/1Oe83LvdOrkFn2o2mvny)
{% endtab %}
{% endtabs %}

You will see a list of available speech options.

Once you select a speech option, the system will use that language and model for voice input (when you use the microphone icon in the chat box) and reading responses back to you.


# Chat Filter

The Chat Filter feature allows users to narrow chatbot responses to specific parts of the knowledge base by selecting predefined fields and values.

> Note: This feature is not available on mobile devices.

#### 1. Purpose

The **Filter feature** helps you narrow down chatbot answers to a specific subset of data.\
By applying filters, you can ask questions that only return results matching your chosen criteria — for example, by *region*, *topic*, or *time period*.

:warning:T*his feature may not be available in all projects— it appears only in GLChat instances where filtering has been enabled by the administrator.*

***

#### 2. When to Use Filters

Use filters before sending a question when:

* You want the project to focus on a particular category, group, or dataset.
* You need consistent filtering for several follow-up questions in the same conversation.

***

#### 3. Opening the Filter Panel

1. In your chat window, click the **Filter** button on the bottom-left corner of the chat input box.\
   ![](/files/RJ2CluP6kHpfprQ7q7tf)
2. A **Filter panel** opens on the right side of the screen, showing available fields (for example, *Geography*, *Speaker Type*, *Therapy Area*, *Date Range*, etc.).\
   ![](/files/VYViFoNErNgOfVMU4IzD)
3. Click the + sign to apply the filter fields.
4. Each field can be used to refine what information the chatbot looks at when generating answers.

***

#### 4. Applying Filters

1. In the Filter panel, review the list of fields.
2. For each field you want to use:
   * Select a value from the dropdown or enter text, depending on the field type.
   * Some filters let you choose **one value**, others allow **multiple selections**.\
     ![](/files/DJMCaMik3W0B9TPw6DwS)
3. Click **Apply**.
4. Once applied, a small status label will appear under the chat input (for example: *1 Filter Selected* or *2 Filters Selected*).

***

#### 5. Sending Messages with Filters

1. After applying filters, type your message in the chat box.\
   ![](/files/fyBCGvKpd9yRNbaAlO2D)
2. Press **Enter** or click **Send**.
3. The chatbot will respond using only the data that meets your selected filter conditions.

***

#### 6. Reviewing or Changing Filters

* To **see** which filters are active, click the **Filter** button again or check the label near the chat input.
* To **edit** a filter, open the Filter panel, change the value(s), and click **Apply** again.
* To **remove a filter**, click the trash icon next to a field.
* To **remove all filters**, click **Remove All** and then **Apply**.


# Task Scheduler

## Schedule a Task

You can schedule a task in GLChat. The tasks can vary, such as:

* Send a prompt to LLM
* Send instructions to AI Agent
* Perform a search (Web/Connector)
* Conduct deep research

{% tabs %}
{% tab title="Website" %}
To create or schedule a task, type a prompt in the chat input box. After selecting one of the tasks above, use the + Button and choose the schedule message.

<figure><img src="/files/hrrLNZDW2KeWFU5FkSsf" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
To create or schedule a task, type a prompt in the chat input box. After selecting one of the tasks above, use the chevron up Button and choose the schedule message.

![](/files/uDjRVPwp38NhYUL0oLOn)<br>
{% endtab %}
{% endtabs %}

**Note**: Currently, you cannot schedule a task for a prompt that contains a file attachment.

There are several types of schedules:

* Once
* Interval (minutes)
* Interval (hours)
* Daily
* Weekly
* Weekdays
* Monthly
* Yearly

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/7wfxmpW1tn7OYhmsQ9ZH" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/FpRCVClm264YqbvmLER2" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/0xidDZjHmASdVizPbc32)
{% endtab %}
{% endtabs %}

When sending an email notification, use the "Notify Others" option to add external email recipients or users from your organization.

Once you are firm with your choice, hit the Save button. A toast message informing you that the task is already created will appear, along with a link to open the Task List.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/9SCgsmJl2WSOYKpxHXLn" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/U81aehCiNktjuCVJg4tw)
{% endtab %}
{% endtabs %}

Once the task meets the scheduled time, it will be executed as a new conversation.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/gBVxJAFg9Qx2dsAZRrIr" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/vQopjObNFELa1f119vU3)
{% endtab %}
{% endtabs %}

## View Task List

To view task lists, open the toast notification that appears after scheduling a task, or simply go to the "Schedule Task" menu in the profile section at the bottom left.

{% tabs %}
{% tab title="Website" %}

<div align="left"><figure><img src="/files/g7ELyUdgynDFtLFmFqFo" alt=""><figcaption></figcaption></figure></div>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/80At4JV9dSwr9njMvuoU)
{% endtab %}
{% endtabs %}

This menu includes two tabs:

**Scheduled**

* Displays all created tasks and their execution status.
* Options available: Set task to inactive, edit, or delete a task.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/ozxsze0f7aBTsGq5nQ94" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/wPCHJLnE1ZixSDrQWugZ)
{% endtab %}
{% endtabs %}

**Completed**

* This section shows all tasks that have already been executed.
* You can also click the completed task to go to the conversation containing the scheduled task.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/8mOsJyib8wfYisB6nOJZ" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/TgBXMSCXss41R29bvq05)
{% endtab %}
{% endtabs %}

## Edit Task

To edit a task, select "Edit" from the Task List. This will redirect you to a page where you can modify the task or its prompt.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/WVY7EWEwQtJP4v2Ef6yE" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/codpGz3mMYzXyMJxlp5g)
{% endtab %}
{% endtabs %}

Note, to change the schedule, simply click the link "Change schedule here".

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/7vp051oVsKmaDvIjJxL2" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/e20TGlUttyq3rMI3MkFv)
{% endtab %}
{% endtabs %}

Once you are done with the edit task, click the check button. A toast message will appear informing you that the adjustment was successful. You can check the task via the Task List menu.


# Chat with Connector

#### Overview

Chat with Connector allows users to interact with connected applications directly from GLChat. With this feature, users can ask questions, retrieve information, and perform supported actions using connectors such as Gmail, Google Calendar, GitHub, and other applications that have been enabled for the organization.

Connectors help users work across different tools without leaving the chat experience. For example, a user can ask GLChat to summarize recent emails, check calendar availability, search GitHub issues, or combine information from multiple connected applications.

<figure><img src="/files/ZnVVsV9CnqVB3BqMMntB" alt=""><figcaption></figcaption></figure>

***

### How It Works

The connector options shown in GLChat depend on the connectors configured by the organization admin. Users can only access connectors that are available in their organization and have been authenticated by the user.

Once a connector is connected, users can ask GLChat to interact with the application using natural language.

For example:

```
Summarize my latest unread emails from Gmail.
```

```
Check my calendar availability tomorrow afternoon.
```

```
Find open GitHub issues related to login error.
```

```
Create a calendar event based on the meeting details from my email.
```

GLChat will use the selected connector capability to process the request. Depending on the connector, GLChat may retrieve information, summarize content, search records, or perform supported actions.

***

### Selecting a Connector

To use Chat with Connector:

1. Open the message input menu.
2. Select **Connector**.
3. Choose one or more available connectors.
4. Enter your instruction in the chat box.
5. Send the message.

The list of available connectors may vary depending on the organization configuration and user authentication status.

<figure><img src="/files/rbzYCJNNtDrYHSVQfUQc" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/tTEIt4ESut49aQK79kBo" alt=""><figcaption></figcaption></figure>

***

### Using Multiple Connectors

Users can select more than one connector for a single request. This is useful when the task requires information or actions across multiple applications.

Example:

```
Check my calendar availability tomorrow afternoon, and send the information to client
```

In this example, GLChat may use Google Calendar to check the follow-meeting schedule and Gmail to send the email.

<figure><img src="/files/3bYdQkeG8kQvMQYKoaGu" alt=""><figcaption></figcaption></figure>

***

### Supported Capabilities

Connector capabilities depend on each application and how the connector is configured. In general, users may be able to:

* Ask questions based on connected application data.
* Search or retrieve relevant information.
* Summarize content from connected applications.
* Perform supported actions, such as creating drafts, scheduling events, or searching issues.
* Combine multiple connectors in one task.

Some actions may require confirmation or additional user input before they are completed.

***

### Authentication and Access

Before using a connector, users may need to authenticate their account. Authentication ensures that GLChat can access the connected application based on the user’s permission.

Access is also controlled by the organization configuration. If a connector is not available, it may mean that the connector has not been enabled by the organization admin or the user has not completed authentication.

***

### Example Use Cases

#### Gmail

```
Summarize unread emails from today.
```

```
Find the last email from Sarah and draft a reply.
```

#### Google Calendar

```
Check my availability this Friday afternoon.
```

```
Schedule a 30-minute meeting tomorrow at 10 AM.
```

#### GitHub

```
Find open issues related to SSO integration.
```

```
Summarize recent pull requests in this repository.
```

#### Multiple Connectors

```
Check my email for the meeting request, then create a calendar event based on it.
```

```
Find GitHub issues related to the latest client feedback from Gmail.
```

***

### Notes

* Available connectors depend on organization settings.
* Users may need to authenticate each connector before using it.
* Connector actions depend on the supported capability of each application.
* Some requests may require more specific instructions if the connector needs additional context, such as date, repository, recipient, or calendar details.
* Users can select multiple connectors when the task involves more than one application.

***


# Smart Mode

#### Overview

Smart Mode helps users interact with GLChat more easily by allowing them to ask questions without manually choosing a specific mode first.

Instead of selecting whether to use LLM, Search, Deep Research, AI Agent, or Chat to Connector, users can simply type their prompt. GLChat will automatically choose the most suitable mode based on the request and the available configuration.

This is useful for users who want a faster and simpler experience, especially when they are unsure which mode should be used.

<figure><img src="/files/MqDhN1UQJS5A3Kc6BseK" alt=""><figcaption><p>Smart Mode On</p></figcaption></figure>

***

#### How Smart Mode Works

When Smart Mode is enabled, GLChat analyzes the user’s prompt and automatically decides the best way to respond.

For example:

* If the prompt is a general writing or brainstorming request, GLChat may use **LLM Mode**.
* If the prompt requires recent or external information, GLChat may use **Search**.
* If the prompt requires broader research and comparison, GLChat may use **Deep Research**.
* If the prompt needs to interact with an available connector, GLChat may use **Chat to Connector**.
* If the prompt requires a more complex workflow, GLChat may use an **AI Agent**.

With Smart Mode, users do not need to understand the differences between each mode before asking a question. GLChat handles the mode selection automatically.

***

#### Key Benefits

**Ask First, Choose Less**

Users can focus on writing their question or instruction instead of deciding which mode to use.

For example, instead of manually selecting Search or Deep Research, the user can simply ask:

> “Find the latest regulation update and summarize the key points.”

GLChat will determine the most appropriate mode based on the prompt.

***

**Automatic Mode Selection**

Smart Mode can route the prompt to the most relevant capability available in the project, such as:

* **LLM Mode** for general answers, drafting, rewriting, or brainstorming.
* **Search** for current or external information.
* **Deep Research** for deeper research, comparison, or multi-source synthesis.
* **AI Agent** for internal workflows or more complex tasks.
* **Chat to Connector** for direct connector-based actions or queries.

The available options depend on the project and organization configuration.

***

**Uses the System Default LLM for LLM Mode**

When Smart Mode routes a prompt to LLM Mode, GLChat will use the system default LLM configured for the project.

The system default LLM follows the first LLM shown in the LLM selection list.

This means users do not need to manually select a model when using Smart Mode. The response will follow the default LLM configuration that has already been set for the project.

***

**Smart Mode Configuration is Managed by Admin**

Smart Mode behavior is controlled from the Admin Dashboard.

Admins can configure whether Smart Mode is available and define which capabilities can be used, such as:

* Whether Smart Mode is enabled or disabled.
* Which modes are allowed.
* Which AI Agents are eligible to be selected.
* Which connectors are available.
* Whether certain capabilities are restricted based on user role or organization policy.

This allows organizations to provide a simpler user experience while still maintaining control over what GLChat is allowed to access or execute.

***

#### Manual Mode is Still Available

Smart Mode is designed to reduce upfront decisions, not remove user control.

Users can still switch to a specific mode manually when needed. For example, if a user specifically wants to use Search, Deep Research, or a certain Agent, they can still choose it directly.

This gives users flexibility: they can rely on Smart Mode for convenience, or manually choose a mode when they already know what they want.

<figure><img src="/files/RC0rMAmnLt558ZInc10c" alt=""><figcaption><p>Smart Mode Off</p></figcaption></figure>

***

#### Example Usage

<table><thead><tr><th width="494">User Prompt</th><th>Possible Smart Mode Selection</th></tr></thead><tbody><tr><td>“Rewrite this message to be more concise.”</td><td>LLM Mode</td></tr><tr><td>“What is the latest update from Bank Indonesia?”</td><td>Search</td></tr><tr><td>“Compare AWS, GCP, and Azure for enterprise deployment.”</td><td>Deep Research</td></tr><tr><td>“Create a GitHub issue for this login bug.”</td><td>Chat to Connector</td></tr><tr><td>“Analyze our support tickets and summarize the top issues.”</td><td>AI Agent</td></tr></tbody></table>

The final selected mode may depend on the project setup, available connectors, available agents, and admin configuration.

***

#### Visibility in the Response

When Smart Mode answers a prompt, GLChat can show which mode was used to generate the response.

This helps users understand how GLChat handled their request and gives better transparency when the system automatically chooses a mode.

For example:

> Answered by: Smart Mode → Search

or

> Answered by: Smart Mode → AI Agent

***

#### When to Use Smart Mode

Use Smart Mode when:

* You are not sure which mode to choose.
* You want to ask quickly without configuring the mode first.
* Your prompt may require different GLChat capabilities.
* You want GLChat to automatically decide the best available path.

For more specific cases, you can still manually select a mode anytime.

***

#### Summary

Smart Mode makes GLChat easier to use by allowing users to type their prompt first and let the system choose the most suitable mode automatically.

It reduces the need to manually decide between LLM, Search, Deep Research, AI Agent, or Chat to Connector, while still allowing manual override when needed. Smart Mode also follows the configuration set in the Admin Dashboard, so organizations can keep control over available modes, agents, connectors, and policies.


# Smart Hub

### Overview

**Smart Hub** allows users to ask questions from one main Project, while GLChat automatically searches the relevant sub-project knowledge bases based on the user’s access.

This helps users find information more easily without manually switching between multiple Projects.

***

### Why Smart Hub is Useful

In GLChat, each Project may have its own knowledge base and access permission.

Before Smart Hub, users who had access to multiple Projects needed to manually switch between Projects to find the right information.

For example:

<table><thead><tr><th width="126.33331298828125">User</th><th width="301.111083984375">Project Access</th><th>Before Smart Hub</th></tr></thead><tbody><tr><td>User A</td><td>Project 1 only</td><td>Opens Project 1</td></tr><tr><td>User B</td><td>Project 2 only</td><td>Opens Project 2</td></tr><tr><td>User C</td><td>Project 3 only</td><td>Opens Project 3</td></tr><tr><td>User D</td><td>Project 1 and 2</td><td>Switches between Project 1 and 2</td></tr><tr><td>User E</td><td>Project 1, 2, and 3</td><td>Switches between Project 1, 2, and 3</td></tr></tbody></table>

This can be inconvenient because users need to know which Project contains the answer before asking.

***

### How Smart Hub Works

With Smart Hub, users only need to open the **Smart Hub Project** and ask their question there.

GLChat will then:

1. Check which sub-projects the user can access.
2. Search only the allowed sub-project knowledge bases.
3. Find the most relevant answer.
4. Return the answer to the user.

<figure><img src="/files/cnDg3sC05W0bcWPis7K5" alt=""><figcaption><p><strong>Smart Hub gives users one place to ask questions while GLChat checks access and searches only the allowed sub-projects</strong></p></figcaption></figure>

***

### Example

Suppose an organization has three Projects:

| Project   | Knowledge Base   |
| --------- | ---------------- |
| Project 1 | HR policies      |
| Project 2 | Finance policies |
| Project 3 | IT support       |

If User E has access to all three Projects, they no longer need to switch between Project 1, Project 2, and Project 3.

They can simply ask from Smart Hub, and GLChat will search the allowed Projects to find the best answer.

***

### Key Benefits

<table><thead><tr><th width="252.22222900390625">Benefit</th><th>Description</th></tr></thead><tbody><tr><td>One entry point</td><td>Users ask from one Smart Hub Project.</td></tr><tr><td>Less manual switching</td><td>Users do not need to move between Projects.</td></tr><tr><td>Easier discovery</td><td>Users do not need to know which Project contains the answer.</td></tr><tr><td>Access-aware search</td><td>GLChat only searches Projects the user is allowed to access.</td></tr><tr><td>Better user experience</td><td>The chat experience becomes simpler and faster.</td></tr></tbody></table>

***

### Important Notes

Smart Hub does **not** give users access to all Projects.

Each user’s access permission still applies. If a user does not have access to a sub-project, Smart Hub will not search that sub-project.

***

### Configuration

Smart Hub is configured by the administrator.

Admins can enable Smart Hub for a Project, link other Projects as sub-projects, and manage whether those sub-projects appear in the Project list.

End users do not need to configure anything manually. If Smart Hub is available, users can simply open the Smart Hub Project and ask their questions.

For more details, please refer to the [Smart Hub Configuration section](/glchat/organization-admin/how-to-guides/project/setup-project/smart-hub-configuration).


# Chat Input and Responses


# Autosuggest and Autocomplete

This feature allows you to receive a suggestion prompt related to the knowledge assigned to the specific AI Agent. You can also directly accept the autocomplete suggestions from the system by using the 'Tab' key.\\

\
Note: The current implementation only supports autosuggest and autocomplete in the New Chat page.

<figure><img src="/files/64JJEyGhkb3G6l15b55B" alt=""><figcaption></figcaption></figure>


# Step Indicator

Capability to show steps / process that run during user send message until the response is generated. It will show the steps involved from backend and also the total duration

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/XzyPx4zBN45cxwMvMV4b" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/ElbjC7VlmUH2tSDsqyg6" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/bQKGbsV0HV59iWlHvCfH" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/FIa3IzPaAeWW5XlpqTQI) ![](/files/aruUkFy0FMxgzRXcoIJ1)

<br>
{% endtab %}
{% endtabs %}


# Message Interaction

## Send Message

You can send a message by click "Send Message" button or hit Enter button once you already type the message in the chat input area

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/5o6KfPX2rcwvhnuKEhwK" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/1wdXMjC3u6WTgEF2X9g1)
{% endtab %}
{% endtabs %}

## Stop Response

If you change your mind and **want to stop the message** you sent earlier, simply click the **Stop Response** button to cancel it.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/LCspZbNtph6KP6KY40mv" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/3aVoBMVvBIBq2Uth6lhc)
{% endtab %}
{% endtabs %}

## Edit Input

If you want to **edit your message**, simply click **Edit**, type your new message, and then click the **Send** button or press **Enter** on your keyboard.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/6rqFWdeRGAkOKQ3IOe4M" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/ROLe2uuKSIdTTUVAvEVz" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/JJKN2Aj1to4FnXOHIKfy) ![](/files/FWEBfOU1FfEO52zdgWjr)
{% endtab %}
{% endtabs %}

You can easily revisit previous or recent questions by clicking the arrow on each message.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/5NolqKwNzg38lQ30cOCW" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/rN8SE8vIGznJqsYHpQWc)
{% endtab %}
{% endtabs %}

## Copy Prompt/Response

You can easily copy the response by clicking the **Copy** button at the bottom of each response. The response is in Markdown format.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/bB6lXbFitiOFAo8Mpd52" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/0TMx9LAb3nzkixNHHQjF)
{% endtab %}
{% endtabs %}

## Thumbs Up / Thumbs Down

You can rate a chatbot response with **Thumbs Up (👍)** or **Thumbs Down (👎)** and optionally add **tags** and a short **reason**. Your feedback is used by the **Memory** system to improve future replies (tone, length, clarity, language).

> ℹ️ Memory updates can be **asynchronous**.

Thumbs Up Button

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/jwtMHPVPo5ZHUANpoTTK" alt="" width="563"><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/ulvc68WCHSy1hcpRHJyb)
{% endtab %}
{% endtabs %}

Thumbs Up Detailed Tags

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/QSFGIDRYiNckAtLIZQoq" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/pJfdjddVjzdASV1LNhVL)
{% endtab %}
{% endtabs %}

Thumbs Down

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/GUELI3QjVZpb8YDd4K5b" alt="" width="545"><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/fCgsMc9ZLxH419KyHvrj)
{% endtab %}
{% endtabs %}

Thumbs Down Detailed Tags

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/xwNThioTcE6diduRgfC7" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/w5yj5yQYf7sYnnJbpi5p)
{% endtab %}
{% endtabs %}

## Regenerate Message

You can easily regenerate the answer by clicking the **Regenerate Message** button and selecting the method you want to use. By **default**, the "**Try Again**" option will **use the most recent method you selected** to generate the message.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/lOpcrlPO8s3fqVZPhz3E" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/IKyBkbcZHYOuY02CQCxi)
{% endtab %}
{% endtabs %}

## Export Response

You can export an individual chat bubble if you want to save or share its contents outside the project.

To do this, click the **three dots (⋮) menu** located below the bubble you want to export. From the menu that appears, choose **Export**, and then select the format you want — for example, **PDF**.

The system will generate the file and allow you to download it to your computer.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/xnYzZVoqvA5CE08p95pF" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/xZ4xjBw3xBr6tyiLbYr9)
{% endtab %}
{% endtabs %}

## Mask Personal Data

The **Mask Personal Data** option allows you to quickly hide sensitive information in the chat response, such as names, emails, or phone numbers, before sharing or exporting the conversation.

When you choose Mask Personal Data, the system will automatically anonymize or mask identifiable details in that specific response (e.g., names, emails, phone numbers).

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/08O8EtwdNRA66p7umf4g" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/XgLk4fJWBcFoF9I5I2PC)
{% endtab %}
{% endtabs %}

## Read Aloud

The **Read Aloud** option lets the system speak the content of the selected chat bubble.

Clicking this option will trigger text-to-speech playback, so you can listen to the response instead of reading it.

**Limitation**: It only supports up to **100 characters (including spaces)** to prevent exceeding the quota.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/Ox8GI7n7KcwzUr7RFanr" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/7fgDALPXpAJhncuW8YKE)
{% endtab %}
{% endtabs %}

## Understanding Chat History (Short-Term Memory)

Chat history refers to the record of past conversations or messages exchanged between users in a chat project. It serves multiple purposes, such as providing context for ongoing discussions, facilitating the retrieval of information, and ensuring accountability in communication.

**Key Benefits**

1. **Contextual Reference**: Maintains a record of past interactions, enabling users to recall previous conversations.
2. **Information Retrieval**: Allows users to search and retrieve information exchanged in past chats quickly.
3. **Accountability**: Provides a transparent log of communications that can be used for verifying information or resolving disputes.

Understanding the significance of maintaining and managing chat history effectively can greatly enhance communication and efficiency within a team.

## Start a separate chat (Branch Message to a New Conversation)

The **Start a separate chat** feature allows you to create a new conversation from any existing message response without losing context. This is useful when you want to explore a different direction, refine a specific answer, or continue a discussion independently while keeping your original chat unchanged. Instead of starting from scratch, you can reuse the existing conversation history and branch into a focused discussion.

To use this feature, begin by chatting as usual. After you receive a response—whether from an LLM reply, an Agent action, or a search result, hover over the message and click the **three-dot (⋯) menu**. Select **Start a separate chat**.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/4gdVdJbPyJW3IUGrvMfp" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/eNvYlef9lFOhkMBrk6wJ)
{% endtab %}
{% endtabs %}

Once selected, it will automatically create a new chat containing the conversation history up to the message you chose. You will be redirected to this new chat, where you can continue the discussion independently.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/RtaR8uGujQieT26CTRwJ" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/fBMErbv4Pio01QjbU2Vb)
{% endtab %}
{% endtabs %}

The original chat remains unchanged, and the new chat behaves like a normal conversation thread. You can rename, manage, or delete it as needed. This feature helps you stay organized, compare different approaches, and prevent long conversations from becoming cluttered—while still preserving important context.


# Quote and Reply

This feature allows you to easily quote specific statements and provide your response for clearer communication. If you have a question or need further clarification about something the bot has said, simply highlight the portion you want to reference.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/gARy3rxStanpbToOIFOk" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
To activate quote and reply action, click the three-dot (⋯) icon located in the bottom of the chat. From the panel that appears, select “Select & Quote”.

![](/files/0TxOUTqwyzt6VWsEMUgX)
{% endtab %}
{% endtabs %}

For example, if the response mentions a specific fact or suggestion, you can quote it by drag or select the statement, "\[quoted statement] Regarding this point, can you explain more about \[your question]?". This helps ensure that your questions are specific and relevant, leading to more efficient and helpful responses.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/heEafrld3dL69re8kRr1" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/xb0ycPfCgDqdbgPmRs7Y) ![](/files/MJmC5A5gcy881l54eV7e)
{% endtab %}
{% endtabs %}

\\


# Display Media Files in the Chat Response Area

Currently, if the answer from the bot contains media files such as images or YouTube previews, the media can be shown inline in the chat area. This will allow you to preview the images or play YouTube videos directly in the chat area. If an image also has a caption, it will be displayed under the image correspondingly.\
\
Note: Currently, the platform only supports displaying images and YouTube video widgets. This feature available only by request

<figure><img src="/files/TVrjkCr757dbfIUCQ62P" alt=""><figcaption><p>Image shown in the chat area</p></figcaption></figure>


# Display Chart in the Chat Response Area

Currently, if the answer from the bot contains chart such as line chart, bar chart, and pie chart, it can be shown inline in the chat area. This will allow you to preview the chart and analyze the data directly.\
\
Note: This feature is available only by request

<figure><img src="/files/511lQ9nzWwOrFe2P2dKt" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/OjkmqZMo7caVRwnpAoL0" alt=""><figcaption><p>Display Pie Chart in the Chat Response</p></figcaption></figure>


# Inline Highlight PDF from Reference

This feature allows you to have an inline preview of a PDF file that serves as a reference source for an answer from the bot. Capabilities introduced in this feature include:

1. Opening a PDF file side by side without the need to move to another tab.

<figure><img src="/files/EHwINLXl6YjL7uvkb0Za" alt=""><figcaption></figcaption></figure>

2. Highlighting specific areas in the PDF based on the context used to answer the question by the bot.

<figure><img src="/files/nXDFUf9CleofuizG9NvZ" alt=""><figcaption></figcaption></figure>

3. Downloading the PDF file

<figure><img src="/files/ITJdNzX36nAfAUdIW2kO" alt=""><figcaption></figcaption></figure>


# Answer Citation

When using **Web Search mode**, the project can provide citations in its answers if references are available. You can simply hover over the citation number and click the link to verify the information from the sources.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/VFuxO2t9uz2mSJyrNQYi" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/yEtcgFSfl4gvepDAJ9iT)
{% endtab %}
{% endtabs %}

<br>


# Sources

You can see the **references or sources** for each response by clicking the Source button at the bottom of the response.

A dialog box on the right sidebar will display a list of sources. You can click each link to verify the information directly from the source.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/rb3Gou1fvERChzWMKeH5" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/CoFKPt8SEySMTvv4wPgx)
{% endtab %}
{% endtabs %}

## SQL Sources

### Overview

GLChat now supports **SQL Sources** for agents using SQL tools.

This feature helps users easily trace and understand how the agent generates answers from database queries.

Users can now see:

* SQL query sources
* Related database results
* Execution references behind the answer

***

### Benefits

| Feature            | Benefit                                         |
| ------------------ | ----------------------------------------------- |
| Source Visibility  | Easier to trace answers                         |
| Query Transparency | Understand how the answer is generated          |
| Easier Debugging   | Faster investigation when answers are incorrect |
| Better Trust       | Users can validate the returned data            |

***

### How It Works

1. Agent generates a SQL query
2. SQL tool executes the query
3. Database returns the result
4. GLChat displays the answer along with SQL Sources

***

### Example

#### User Prompt

> “Berapa kenaikan produksi dalam perbandingan plan dan actual masing-masing powerplant antara tahun 2023 dan 2024 dan visualisasikan dalam bentuk bar chart?”

#### SQL Sources

Users can inspect:

* Executed SQL query
* Related tables
* Retrieved results

This makes answers more transparent and easier to validate.

<figure><img src="/files/zy3Ch4xrGfiNjTM9KGNd" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/jnvPPCRfQeLLG9Vh2Yx4" alt=""><figcaption></figcaption></figure>


# Related Queries

When using Web Search, the chatbot can provide **recommended related questions based on the given context**.

Simply select one of the related questions, and it will automatically fill the chat box.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/FTvcGB0ka6b1N2cXmQKr" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/JfVLlwZR9L3kjP9S5Xag)
{% endtab %}
{% endtabs %}

Then you need to click the **Send** button or press **Enter** on your keyboard.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/26zvEEtYWHJUMVo8bmHA" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/lMGJONDSJftmQ8KMkAyS)
{% endtab %}
{% endtabs %}


# Generate Clickable Mindmap from Document

In GLChat, once you already upload document file to a conversation, you can create clickable mindmap from it. Below are the steps to generate the clickable mindmap:

1. Start a conversation in GLChat
2. Upload a document file (PDF / DOCX) to the conversation and ask something (e.g explain about this document)
3. Once the document already uploaded, you can go to the Chat Control in right sidebar

<figure><img src="/files/QERJE35773x8ilao38Uo" alt=""><figcaption></figcaption></figure>

4. Click the File Tab, you should see the file that uploaded

<figure><img src="/files/xdfplLZOYySQtATnb2Ie" alt=""><figcaption></figcaption></figure>

5. Click the Generate Mindmap button (right side of the file name)

<figure><img src="/files/p1Wj8054Q61yUjgspwS8" alt=""><figcaption></figcaption></figure>

6. It will show the Clickable Mindmap for the document

<figure><img src="/files/Auc2phRY4HB10OXRTmGe" alt=""><figcaption></figcaption></figure>

7. To expand or collapse, hover the pointer to edge of a node, and it will show an option to collapse or expand as in example below

<figure><img src="/files/R9A4UJBU8jBS3zvbMe19" alt=""><figcaption></figcaption></figure>


# Question Library

A centralized repository of questions that combines admin configured prompts and or auto-generated FAQ, for guided conversation starters.

Questions in question library are organized by categories for easier navigation and discovery. In addition, the Question Library appears contextually in the chat input area, giving users quick access even after multiple turns in an ongoing conversation

{% tabs %}
{% tab title="Website" %}
List of questions and Question Library in home page

<figure><img src="/files/ndU2cxS6aIUFNvK09yMW" alt=""><figcaption></figcaption></figure>

List of questions under Question Library

<figure><img src="/files/OK0vg0sSFueA2noKzCYh" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
List of questions and Question Library in home page

![](/files/eflrAKi9SQKcBbQlI8Ns)

List of questions under Question Library

![](/files/jayBWLpbZ7PUwtEpq2dl)
{% endtab %}
{% endtabs %}

When clicked, the question can directly processed, or editable, depends on the configuration in the admin dashboard.


# Show Activity and Thinking

This feature displays intermediate steps or background actions performed by the AI while it processes a request. For example, users can see when the system is analyzing a query, fetching data from multiple sources, or reasoning through multi-step instructions. It helps users understand how the AI arrives at its conclusions and provides reassurance that the model is actively working on their request.\
\
Currently, **Show Activity and Thinking** is available in several features including **AI Agent**, **Deep Research**, and **Web Search,** where it reveals activities such as retrieving documents, summarizing results, or planning next actions.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/TbWbsLXqX6GA1DVZAKg7" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/ErjnllFkok9xGrLwJ9Bo" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/11WKkhPB950eq85pYdqh) ![](/files/Ar4iwuESn1hSSrAbkWwc)
{% endtab %}
{% endtabs %}


# Markdown Support

With Markdown Support, GLChat can present responses in a structured and visually organized format making technical explanations, multi-step instructions, or data presentations easier to follow. Users can view formatted text including **bold**, *italic*, bullet lists, numbered steps, and even code snippets or tables directly within the chat interface.\
\
This feature enhances readability and user experience, particularly for documentation, coding assistance, and analytical reports. Markdown rendering is supported across both the **chat interface** and **AI Agent responses**, ensuring consistent formatting whether users interact through the web app or integrated platforms.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/qXx7On8sIodB5zfrcYLs" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/ISpixGUDZd8h8YbRuprq" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/9CRMKtCYnkEmu4W6zHSD)
{% endtab %}
{% endtabs %}


# HITL

Human-In-The-Loop

The GLChat UI allows users to interact directly with AI agents. When the agent encounters a step requiring human validation, it triggers a **HITL** confirmation prompt.

#### How It Works

1. User sends a message to the agent.
2. AI agent processes and pauses when it reaches a critical action that already configured in during create / edit agent. Please refer to[ HITL Setup During Create / Edit Agent section](https://gdplabs.gitbook.io/glchat/admin-dashboard/how-to-guides/ai-agent#hitl-setup-during-create-edit-agent).
3. GLChat UI displays **Approve / Reject / Skip** buttons inline in the conversation.

<figure><img src="/files/cRt4qw5tKAEycFV9ZwpR" alt=""><figcaption></figcaption></figure>

<br>

4. Upon selection:

* The system processes the response.
* The workflow resumes or stops accordingly.
* A confirmation message appears in chat (e.g., “Decision submitted successfully”).

<figure><img src="/files/11xmuPu3rfnO3qLypb4j" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/EMX152zCV0VyTNj8ZUKf" alt=""><figcaption></figcaption></figure>

5. If the user does not respond before timeout, the agent automatically skips the step and remaining steps continue.


# Slash and Mention Command

Enable users to perform quick actions, invoke AI Agents, Search Deep Research, or control chat sessions directly from the chat input using intuitive /commands and @mentions. This improves productivity, reduces UI friction, and aligns with modern chat-based interaction standards seen in leading AI assistants and collaboration platforms.

Both / and @ command available via Chat Input

## Slash (/)

Commands that are triggered when a user types "/" in the chat input will initiate an action. Currently, these commands only work when the slash is the first character in the input.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/uK76NZzIEDCRFsYP6FT0" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/iNV8XOUHKbHmmrYRFFew)
{% endtab %}
{% endtabs %}

Below are the list of action that can be performed using / command:

| Slash Command   | Description                                             | Conditions                                                                                  |
| --------------- | ------------------------------------------------------- | ------------------------------------------------------------------------------------------- |
| /upload         | Open file upload dialog                                 | Only visible when enabled. Requires file-upload feature to be active.                       |
| /settings       | Open the settings drawer                                | Only visible when enabled.                                                                  |
| /tools          | Open tools dropdown                                     | Only visible when enabled. Mirrors existing Tools button.                                   |
| /model          | Open model selector dropdown                            | Only visible when enabled. Does not switch model automatically.                             |
| /files          | Show list of uploaded files in the current conversation | Only visible when enabled. Requires file history, active conversation and uploaded files.   |
| /rename         | Rename the current conversation                         | Only visible when enabled. Only works after the first message when conversation is created. |
| /share          | Open the conversation-sharing modal                     | Only visible when enabled. Only shown if sharing is allowed & user has correct RBAC.        |
| /anonymize on   | Enable data anonymization                               | Only visible when enabled. Toggles anonymization to ON.                                     |
| /anonymize off  | Disable data anonymization                              | Only visible when enabled. Toggles anonymization to OFF.                                    |
| /switch project | Switch to another project                               | Only visible when enabled. Shows list of available project for the Organization.            |

## Mention (@)

When a user types @, a set of commands is activated, enabling modes like AI Agent, Web/Search Connectors, and Deep Research. Only one mention can be executed at a time.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/lkkkBt72xDk62ihQLMJh" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/aFIkYti9m8hahZLdv2fm)
{% endtab %}
{% endtabs %}

Below are the list of mentions command:

| Mention Command | Description                                                     | Example                                                                      |
| --------------- | --------------------------------------------------------------- | ---------------------------------------------------------------------------- |
| @WebSearch      | Fetch latest web content or answers from real-time sources.     | @WebSearch find latest news about blockchain                                 |
| @DeepResearch   | Compare, summarize, or derive insights from multiple documents. | @DeepResearch analyze latest trend about generative AI                       |
| @\[AgentName]   | Route query to a named internal or custom agent.                | @EmailAssistant prepare draft email about negotiation new proposal next week |
| @\[Connector]   | Fetch content from systems like Drive, Notion, Jira, etc.       | @Calendar list down my today's meeting                                       |


# Agent with User Authentication

This guide explains how to create an **agent that requires user authentication** when using tools or MCP connectors.

Currently, this feature is **only supported for tools/MCPs integrated via GL Connectors that enabled within your organization**.

For connector configuration, see: <https://gdplabs.gitbook.io/sdk/gl-ai-agent-package/guides/agents#gl-connectors-token>

***

## Configuration

To enable user authentication, set `user_authentication: true` in the tool or MCP configuration.

{% code title="agents.yaml" %}

```yaml
tool_configs:
  <tool-id>:
    user_authentication: 'true'
    
mcp_configs:
  <mcp-id>:
    user_authentication: 'true'
```

{% endcode %}

<figure><img src="/files/i28Yekv2I1WqFYFehGxj" alt=""><figcaption></figcaption></figure>

When enabled, the agent will require the user to authorize the connector before executing the tool.

***

## How It Works

1. The agent detects which connectors require authentication.
2. If a connector is not authorized, the agent will request authorization.
3. The agent waits until **all required connectors are authenticated**.
4. After authorization is completed, click **Retry** to run the query again.

The agent will then execute the task using the **user's authenticated account**.

***

## Example

Example: **Email Assistant Agent using Gmail**

{% stepper %}
{% step %}

### User request

User asks the agent:

```
Send an email to john@example.com saying the meeting is rescheduled to tomorrow.
```

{% endstep %}

{% step %}

### Connector detection

The agent detects that **Gmail authorization is required**.\
\
Click the Connectors button, it will redirect you to the list of connectors.\
![](/files/HHLQXU3b21EIiqDX4Ck7)
{% endstep %}

{% step %}

### Prompt user

The user is prompted to authorize the Gmail connector.\
Click **Connect** in the required connector and do the authorization.\
![](/files/yMGQ6XYT1d97yIQVmwTM)
{% endstep %}

{% step %}

### Retry

![](/files/QrJwOQJ4OiNAyE3vjjSr)\
After authorization, close the Connectors window and click **Retry** in conversation window.\
![](/files/yKxA0viq8R4nhrG4qfsw)
{% endstep %}

{% step %}

### Execute

The agent sends the email using the **user's Gmail account**.\
![](/files/e1HmOblJTTi7I01mXrjL)

\
![](/files/tA6yeN8H8LOmr34L3ocy)
{% endstep %}
{% endstepper %}

***

{% hint style="info" %}
Notes

* Authentication is required **once per connector** unless the token expires or is revoked.
* If multiple connectors require authentication, **all must be authorized before the agent runs**.
  {% endhint %}


# AI Agent As Default

To set an AI Agent as the default for interaction:

* Use the Admin Dashboard to enable this setting.
* Once enabled, the project will default to this AI Agent for interactions.

#### Important Notes

* When "AI Agent as Default" is active, users are limited to interactions supported by the AI Agent.
* Features like Search or Switch LLM are not available in this mode.


# Display Mermaid Diagram

Renders Mermaid diagrams directly in the chat response for easier visualization

Enables direct rendering of Mermaid diagrams within chat responses for enhanced visualization.\
Assists users in comprehending flows, organizational structures, and processes intuitively without the need to interpret or read raw diagram code. This facilitates clearer communication and understanding, making complex systems more accessible and comprehensible to the user. By utilizing visual representations, it minimizes potential misinterpretations and provides a more engaging and effective way to convey information.

<figure><img src="/files/1r2gffzGLHQMYd8QuWKm" alt=""><figcaption></figcaption></figure>

You can either copy the raw mermaid diagram code or download the rendered mermaid diagram.

<figure><img src="/files/KLA7LUSPMPRGliGLHfVa" alt=""><figcaption></figcaption></figure>


# Copy and Download Tables as Markdown

GLChat now supports copying and downloading tables from chat responses in Markdown format.

This feature helps users reuse table outputs more easily in documentation, reports, GitBook pages, GitHub issues, or other Markdown-supported platforms.

### What You Can Do

When a chat response contains a table, you can now:

| Action               | Description                                                                   |
| -------------------- | ----------------------------------------------------------------------------- |
| Copy as Markdown     | Copy the table content in Markdown format and paste it into another platform. |
| Download as Markdown | Download the table as a Markdown file for later use or documentation.         |

### Why This Is Useful

Previously, users needed to manually reformat tables when moving them from chat responses into documentation or reports. With this enhancement, table outputs can be reused more efficiently while preserving a structured Markdown format.

This is especially useful for:

* Creating documentation from AI-generated tables
* Preparing reports or analysis summaries
* Sharing structured outputs in GitBook or GitHub
* Reducing manual formatting effort

### How to Copy a Table as Markdown

1. Ask a question that generates a table response.
2. Hover over or open the table action menu.
3. Click the **Copy** icon.
4. Paste the copied content into a Markdown-supported editor or platform.

### How to Download a Table as Markdown

1. Ask a question that generates a table response.
2. Hover over or open the table action menu.
3. Click the **Download** icon.
4. The table will be downloaded in Markdown format.

### Notes

* This feature is available for tables generated inside chat responses.
* The copied or downloaded output follows Markdown table format.
* The result may still need minor adjustment depending on the platform where it is pasted.

<figure><img src="/files/cgvs6NrZ7eiadMwruQT3" alt=""><figcaption></figcaption></figure>


# HTML Preview in GLChat

Preview HTML output generated by the LLM directly beside your conversation, review it in full screen, and download the HTML code when needed.

### Overview

When GLChat generates HTML code, you can preview the result side-by-side with the conversation. This allows you to quickly review the visual output without copying the code into another tool or opening a separate editor.

This feature is useful for reviewing generated UI prototypes, simple web pages, forms, layouts, or other HTML-based outputs directly inside GLChat.

### Key Capabilities

#### Preview HTML Side-by-Side

If the LLM response contains valid HTML code, GLChat can display a live preview beside the chat conversation.

This helps you compare the generated code and the visual output in the same workspace, making it easier to validate the result and provide follow-up instructions.

<figure><img src="/files/pLpov8qw3NL57yhF3Syk" alt=""><figcaption></figcaption></figure>

#### Maximize the Preview

You can maximize the HTML preview to view it in full screen.

This is helpful when the generated output contains a larger layout, interactive elements, or detailed UI components that are easier to review with more screen space.

<figure><img src="/files/HTk0B8T2S93XSNel3JVe" alt=""><figcaption></figcaption></figure>

#### Download the HTML Code

You can download the generated HTML code for further use, testing, or sharing with your team.

After downloading, the file can be opened locally in a browser or handed over to the relevant team for further refinement.

### Privacy and Security Reminder

Since the HTML content is generated by an LLM, please avoid entering real, sensitive, or confidential data during preview.

Examples of data that should not be entered during preview include:

* Real customer information
* Passwords or credentials
* Personal identification data
* Financial information
* Internal confidential business data

Use dummy or sample data when testing the generated HTML preview.

### Interactive Preview Limitation

Some HTML previews may include interactive behavior such as forms, buttons, scripts, or dynamic UI components.

{% hint style="info" %}
For privacy and security reasons, additional setup or configuration may be required before full interactive preview is enabled. This helps ensure that generated content is handled safely, especially when scripts or user input are involved.
{% endhint %}

<figure><img src="/files/26iRQeqeXPOJDAFZ3zk3" alt=""><figcaption></figcaption></figure>

### Recommended Usage

Use the HTML preview feature to quickly validate generated UI or web-based content before deciding whether to refine, download, or share it.

For best results:

* Start with a clear prompt describing the HTML output you need.
* Review the side-by-side preview.
* Maximize the preview if the layout needs more space.
* Use dummy data during testing.
* Download the HTML code if you need to continue editing or share it externally.


# Interactive Agent Responses

Agents in GLChat can display interactive components such as forms, cards, buttons, or guided actions when configured with A2UI.

#### Overview

GLChat supports interactive Agent responses through **A2UI**. With this capability, an Agent response can be rendered as an interactive UI component inside the chat, instead of being shown only as plain text.

This allows users to interact with the Agent output more naturally, such as filling in forms, selecting options, reviewing structured information, confirming actions, or continuing a guided flow directly from the chat interface.

For example, instead of asking the user to reply manually with several details, the Agent can show a structured form or selection component. The user can then complete the interaction directly in GLChat.

<figure><img src="/files/ryPfpcCxPWVhQCMddObQ" alt=""><figcaption></figcaption></figure>

This helps users interact with Agent outputs in a more structured and visual way.

#### What Makes It Interactive

Interactive Agent responses allow the Agent to present information and collect user input using UI components.

Depending on the Agent configuration and supported components, the response may include:

* **Forms** for collecting structured input from users.
* **Buttons** for triggering actions or confirming the next step.
* **Selections** for choosing one or more options.
* **Cards or lists** for displaying structured information.
* **Tabs** for grouping related information.
* **Modals or confirmation dialogs** for review and approval steps.
* **Date, checkbox, slider, or text input fields** for guided data entry.

This helps reduce back-and-forth conversation when the Agent needs specific information from the user.

#### Example Use Cases

Interactive Agent responses are useful when the conversation requires more than a simple text answer.

Common use cases include:

* **Form submission**\
  The Agent can ask the user to complete a form, such as request details, issue information, or additional data needed to continue a workflow.
* **Confirmation step**\
  The Agent can show a summary and ask the user to confirm before continuing with an action.
* **Option selection**\
  The Agent can present several choices, such as categories, priorities, available actions, or recommended next steps.
* **Structured result display**\
  The Agent can show results in a card, table-like layout, list, or grouped section so the information is easier to scan.
* **Guided workflow**\
  The Agent can guide users step by step using interactive components, instead of relying only on long text instructions.
* **Action-oriented response**\
  The Agent can provide a button or input component when the user needs to continue an action directly from the chat.

#### How It Works

When a user interacts with an Agent, the Agent may return an A2UI-compatible response.

GLChat will detect the response and render it as an interactive component in the chat interface, as long as the component is supported.

The rendered result depends on:

* The A2UI components supported by the Agent.
* The rendering capability supported by GLChat.
* The Agent configuration.
* The response format generated by the Agent.
* Any additional behavior implemented on top of A2UI in GLChat.

If the Agent returns a supported interactive response, GLChat will render it directly. If the response is not supported, additional implementation or configuration may be required.

#### Supported Interaction Patterns

The exact component support may depend on the current GLChat implementation and Agent configuration. In general, interactive Agent responses may support patterns such as:

| Interaction Pattern | Description                                                        |
| ------------------- | ------------------------------------------------------------------ |
| Text input          | User enters information into a field.                              |
| Selection           | User chooses from available options.                               |
| Button action       | User clicks a button to continue or trigger an action.             |
| Confirmation        | User reviews information before proceeding.                        |
| Structured display  | Agent presents information using cards, lists, or grouped layouts. |
| Guided flow         | Agent guides the user through multiple steps using UI components.  |

#### Key Benefits

Interactive Agent responses help make Agent interactions more structured, visual, and easier to complete.

Key benefits include:

* Reduces long text-based instructions.
* Makes Agent responses easier to understand and act on.
* Helps users provide structured input with less ambiguity.
* Improves guided workflows inside GLChat.
* Supports richer Agent experiences beyond plain text.
* Enables clearer confirmation and review steps before an action is continued.

#### When to Use

Use interactive Agent responses when the Agent output is better presented as a UI component instead of regular text.

This is recommended when:

* The Agent needs the user to provide several pieces of information.
* The user needs to choose from predefined options.
* The Agent needs confirmation before continuing.
* The response contains structured information that is easier to scan visually.
* The interaction is part of a guided workflow.
* The Agent output is action-oriented and requires user input.

For simple explanations, summaries, or general Q\&A, a regular text response may still be sufficient.

#### Configuration Notes

To use interactive Agent responses, the Agent needs to be configured to return an A2UI-compatible response format.

The experience may vary depending on:

* Which A2UI components are enabled.
* Which components are supported by GLChat.
* How the Agent prompt or tool response is configured.
* Whether the required interaction behavior has been implemented in GLChat.

For example, an Agent may be able to return a form-like response, but the final behavior depends on whether the form components and submission handling are supported in GLChat.

#### Technical Reference

For the list of supported A2UI features and components, refer to the dedicated [**SDK A2UI** technical documentation page](/glchat/sdk-a2ui/overview)

#### Note

Interactive Agent responses depend on both the A2UI component support and the GLChat rendering implementation.

If a component or behavior is not supported yet, additional implementation or configuration may be required before it can be used in GLChat.


# Settings

To access your account settings, click on your **profile menu** in the bottom-left corner of the application. This menu will expand to show several options. Select **Settings** from the list to open the Settings window.

From here, you can access different tabs such as:

* **General** – Adjust basic application preferences (like language or theme).
* **Profile** – View your username, organization, email, and phone number details.
* **Connector** – Manage integrations like Google Drive, Gmail, or other connectors.
* **Shared Conversation** – Manage any chats you have shared with others.

This is where you can manage and review your personal and organizational settings in one place.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/joe910VyHYhoeh6TXO1D" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/ITyuU4YTnncb3n8WOErn)
{% endtab %}
{% endtabs %}


# Memory Management

GLChat’s Memory feature lets the chatbot remember important facts and preferences across conversations—so follow-up chats feel natural and personalized.

### 1. Turn Memory On or Off

1. In the GLChat app, **open your profile menu** (bottom-left corner).
2. Choose **Settings → Personalization**.
3. Go to the **Memory** section.
4. **Toggle the switch** to enable or disable memory.\
   \&#xNAN;*When memory is off, no new information is stored.*

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/IrVZW0wCEJSqxmaigb64" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/zIuZIZbKJdFkWJbIZcYs)
{% endtab %}
{% endtabs %}

### 2. How Memory Extraction Works

* During any conversation in an application, **type normally**.
* GLChat automatically **extracts key information** (for example: project needs, user preferences) and saves it as memory items.
* These items can later be used to personalize answers.

### 3. Manage Your Memory

Go to **Settings → Profile → Manage Memory** to:

* **View** all stored memory items.
* **Search** to quickly locate a specific item.
* **Delete** individual items using the trash icon.
* **Clear All** memories with the **Clear Memory** button if you want to start fresh.
* **Favorites** keep your most important knowledge at the top of the list.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/U8aRQGTcEBoqiPQR19zf" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/8YtPYQGSryx1RprmzHn1)
{% endtab %}
{% endtabs %}

### 4. Privacy & Control

* All memory is stored **within your organization’s secured environment**.
* You can **disable or clear memory at any time**, ensuring full control of your data.

See also: [Thumbs Up and Thumbs Down](/glchat/application-ui/how-to-guide/chat-input-and-responses/message-interaction#thumbs-up-thumbs-down)


# Change Theme

To change the appearance of the application, open the profile menu at the bottom-left corner of the screen and select **Settings**. A settings popup will appear. From the sidebar on the left, choose **General**.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/CcsIxEtCMOEv7664cQSM" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/wubH81f0AER8wsoHpmrJ" alt="" width="370"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

In the General tab, you will see the Theme option. Select the theme you want — Light, Dark, or System Default — and the interface will update instantly.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/7czAtRX5fQnFla1Myfjm" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/I43pjQknNzXBRvZx9Oh0)
{% endtab %}
{% endtabs %}


# Change Language

To switch the application language, open the profile menu in the bottom-left corner of the screen and choose **Settings**. A settings popup will appear. From the sidebar on the left, select **General**.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/jtNUvs800ZREZPQDgFXW" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/RuJ3u8ZifMcv7EsB0H1t)
{% endtab %}
{% endtabs %}

Inside the General tab, look for the **Language** option and choose either **English** or **Indonesian**. The interface will immediately update to the selected language so you can continue using the application without refreshing the page.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/CjiAsLmqltN8MR0trk36" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/ZqM7a77U63Ug4lNrrUSA)
{% endtab %}
{% endtabs %}


# Change Profile Setting

The **Profile** page in Settings allows you to view and manage your personal account information. You can access it by clicking **Settings** (usually from the top-right corner menu) and then selecting **Profile**.

On this page, you can see:

* **Username** – Your display name in the application.
* **Organization** – The organization or workspace you belong to (this may be set by your admin and usually not editable).
* **Email** – The email address linked to your account.
* **Phone Number** – Your registered phone number, including the country code.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/ai4IC6pVBO2GV3ZLSM5r" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/KO3113ebIQFTIforNICg)
{% endtab %}
{% endtabs %}

This page is primarily for reference, as most of these fields are managed by your administrator. If you need to update your email or phone number, you may need to contact your system admin or follow your organization’s account update process.


# Preference on LLM Model, Anonymize Personal Data and Default Project

This feature allows you to select preferences for the LLM Model, Anonymize Personal Data and Default Project configuration, which will be applied during interactions with the bot. To access this menu, simply go to the profile settings. In the General tab, you will find options for Model Preference, Anonymize Personal Data and Default Project.\
\
"System default" means that the Model Selection, Anonymize Personal Data or Default Project will follow the preset configurations of the system.\
\
"Save last usage" indicates that the last selected configuration (LLM Model, Anonymize Personal Data or Default Project) will be applied to your next chat with the bot.\
\
\*Note: The current implementation saves preferences in the browser's local storage. Therefore, if you switch to a different browser or device, it will revert to the system default settings.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/RFMJQC4g1JeRiFCmUPsa" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/toP7im7RCZ4g90N5g7YW" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/JMGHJT51tgO2tmqlXEuC" alt="" width="224"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}


# Connector as Search Option

Connectors let you search directly within your own tools and data sources (e.g., Gmail, Google Calendar, Google Drive, GitHub) instead of only relying on the web or knowledge base. This helps you quickly find emails, files, events, or code without leaving the chat

## **Prerequisite**

Before using connectors in search, you must connect your identity for each service, currently supported connectors:

* Gmail
* Google Calendar
* Google Drive
* GitHub

Connectors can be linked in two ways:

* **Profile Settings** → Connect your account once and reuse across sessions.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/aXUVTy6oGhYph5axwG60" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/gwG1VgSjQA213Evrpq5d" alt="" width="224"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

* Click **+ Icon,** choose **Tool Dropdown → Search Option** → Connect directly when you select a connector.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/oAvsfQgNIWqK1xzYHTvr" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/ejKsAdJ0ZeQj1izm8ajI" alt="" width="224"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

## How to Use Connectors in Search

1. Open the Tools dropdown in chat input.
2. In the Tool dropdown, choose **Search** and toggle ON the connector(s) you want to search.
3. Enter your prompt and send.
4. Results will be retrieved from the selected connector(s).

## Example of search to Gmail

<figure><img src="/files/9u4Iq08QGTtaun4JmGXw" alt=""><figcaption></figcaption></figure>

## Example of search to Calendar

<figure><img src="/files/9qzqudgFCXe6D2zQoUkI" alt=""><figcaption></figcaption></figure>

## Example of search to Google Drive

<figure><img src="/files/rwH1Ex7TTDIbQUv19Aai" alt=""><figcaption></figcaption></figure>

## Example of search to Github

<figure><img src="/files/lgG0O6G8TeDEeagYQeNh" alt=""><figcaption></figcaption></figure>


# Manage Shared Chat

Manage and organize your shared chat in one place

You can access Shared Chat Management menu in Profile Setting<br>

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/cAuMu3hFpGaYYnYBBJ08" alt=""><figcaption><p>Manage Shared Chat in Profile Setting</p></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/fFawXXfRPt0LYNQbGX6Z)
{% endtab %}
{% endtabs %}

In this menu, you can:

1. View the list of chats you shared before
2. Remove the shared chat link
3. Shortcut to go to the chat room


# Help Center

The **Help Center** provides quick access to support resources directly from within the project

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/sSDJSAZoyQxC45sOyECZ" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/kfIXzygsUkq3ROcMz7Xm)
{% endtab %}
{% endtabs %}

You can open it by clicking your **profile menu** in the bottom-left corner, then choose **Help Center** from the sidebar.

## Report an Issue

This option lets you report a problem, bug, or unexpected behavior in the project.

When you click **Report**, you will be redirected to a form site to fill out details such as what went wrong and how to reproduce the issue. This information helps the support team investigate and resolve problems faster.

## User Guide

This option opens the official documentation or guide for the platform in the Gitbook. It contains step-by-step instructions, explanations of features, and tips for getting the most out of the platform — perfect for both new users and experienced ones who need a quick reference.


# Personalization Prompt, Tone and Style

Enable users to personalize application responses by setting their preferred tone, style, and level of detail, ensuring interactions feel more natural, relevant, and aligned with individual communication preferences

You can access personalization via profile setting

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/kgGJnlh4Rx5M5kNeiBXe" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}

<figure><img src="/files/MPiDnlX81uVyTLj78Db7" alt="" width="375"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

**Note**:

1. Turning off personalization will default to Project settings.
2. Personalization also won't work if the admin configures the Application to disallow it.

When personalization enabled, there are 3 field that you can configure

### Tone

<figure><img src="/files/w0swEUrITkeC9oxoX8mr" alt=""><figcaption><p>Selection of tone</p></figcaption></figure>

There are 4 selections for Tone:

1. **Default**: This is the basic mode, often simple and unadorned, suitable for general purposes.
2. **Friendly**: Engages the reader with a warm and approachable tone. Ideal for content aimed at a broad audience where relatability is key.
3. **Formal**: Employs a professional and structured tone. This is used in official documents, academic papers, or in contexts where a serious and respectful approach is required.
4. **Technical**: Provides detailed and precise information, often full of jargon and specific terminology. Used in manuals, guides, or technical documentation where accuracy and clarity are paramount.

### Style

<figure><img src="/files/9f0g1z6quZyK9YULPm68" alt=""><figcaption><p>Selection of writing style</p></figcaption></figure>

There are 4 selections for Style:

1. **Default**: The standard or pre-set mode presented to users.
2. **Concise**: Information presented in a clear and brief manner.
3. **Explanatory**: Detailed information aiming to clarify complex topics.
4. **Natural**: Human-like, easy-to-understand, and conversational tone.

### Custom Instruction

You can customize the application's behavior by setting specific instructions. For instance, you may choose to have it respond like a helpful assistant, providing detailed and thoughtful answers to inquiries.

Another example, you can configure it to always include emojis in its responses, adding a playful and engaging element to the conversation. Here's an example of personalization using custom instruction and the response:

<figure><img src="/files/RL4IKNlzllIy5q4C27Pz" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/dnUZ9iP1NH1vwbaAIGUU" alt=""><figcaption><p>Response using personalized custom instruction</p></figcaption></figure>

By tailoring these settings, you can enhance user interaction, ensuring the application's responses align with your desired tone and style, thereby improving user satisfaction and experience overall.


# External & Advanced Integrations

<table><thead><tr><th width="63.29901123046875">No</th><th width="169.08660888671875">Feature</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td>Implement Content Safety / Guardrail - 1st Phase</td><td>GLChat Content Safety is a configurable system of input and output controls designed to ensure chatbot interactions remain safe, compliant, ethical, and aligned with organizational policies</td></tr><tr><td>2</td><td><a href="/pages/SP7KvVv2QmQBw1YkhFnT">White Label</a></td><td>Capability to set logo, color theme and several features flag in GLChat</td></tr><tr><td>3</td><td>On Premise Installation</td><td>Deploy GLChat securely within your own infrastructure</td></tr><tr><td>4</td><td>Integration with Model Context Protocol (MCP)</td><td>Capability to allow GLChat to be integrated using MCP technology, whether GLChat as MCP Server or GLChat as MCP Client</td></tr><tr><td>5</td><td>Integration with Agent to Agent (A2A)</td><td>Capability to allow GLChat to connect with various AI Agent platform</td></tr><tr><td>6</td><td>Integration with WhatsApp</td><td>Seamlessly connect with GLChat on WhatsApp for instant support and engaging conversations, right at your fingertips</td></tr><tr><td>7</td><td>Bind GLChat Username (Email) and Phone Number</td><td>Allows the system to associate a user’s email (used in GLChat web interface) with their phone number (used in WhatsApp or other channels).</td></tr><tr><td>8</td><td>Authentication and Register user Whatsapp number in GLChat Whatsapp</td><td><p>Enables user authentication and phone number registration when accessing GLChat via WhatsApp. Upon first interaction, users are verified using a one-time password (OTP) mechanism.</p><p>Once authenticated, their WhatsApp number is registered and linked to their user profile in GLChat, enabling secure, personalized experiences and access control across sessions.</p></td></tr><tr><td>9</td><td><a href="/pages/HrW15tcMrTgw4nABtrRv">Browser Use</a> (via Agent)</td><td>Adds a “Browser” capability that lets GLChat fetch and summarize web content on-the-fly, extending knowledge beyond static knowledge bases.</td></tr><tr><td>10</td><td><a href="/pages/GWXG6RotrRUUc80qIKQG">Guest Mode</a></td><td>Allows users to access and interact with GLChat without requiring authentication or account registration. In Guest Mode, users can immediately start chatting while being restricted to predefined configurations set by the admin.</td></tr><tr><td>11</td><td>NLWeb</td><td>Enhance experience in GLChat by integrating with NLWeb, bridging advanced AI capabilities with intuitive web interactions, ask questions directly to the web in natural language</td></tr><tr><td>12</td><td><a href="/pages/l2ZNRJ3T3W9RSHtOAMoW">Code Interpreter</a></td><td>Enhance interaction in GLChat by integrating with the Code Interpreter, turning complex queries into actionable insights effortlessly, run code and analyze data in real time</td></tr><tr><td>13</td><td><a href="/pages/ms7j76ogf3iVRgZj5nCZ">Knowledge Graph in GLChat</a></td><td>Adds structured relationships between entities to improve GLChat’s understanding, retrieval accuracy, and contextual reasoning. Lays the groundwork for smarter, more connected responses.</td></tr><tr><td>14</td><td>PII Handling in Embedding process</td><td>Introduces an automated redaction layer that detects and masks personally identifiable information before generating embeddings, ensuring data privacy compliance</td></tr><tr><td>15</td><td><a href="/pages/tjrEPHq9mFVzeLyKLySb">GLChat Live</a></td><td>Enhances GLChat's functionality by integrating advanced real-time interaction capabilities through the use of shared microphone and video features. This update allows users to engage in seamless live conversations and video communications, leading to more dynamic and interactive exchanges. This enhancement significantly elevates the responsiveness and engagement level in conversations on the platform.</td></tr><tr><td>16</td><td>Multitenancy</td><td>Allows multiple organizations to run independently within a single GLChat deployment. Each tenant has isolated users, data, and configurations, ensuring privacy, scalability, and efficient resource management</td></tr><tr><td>17</td><td>Agent as Default Mode</td><td>Sets the AI Agent as the primary interaction mode in GLChat. Instead of single-turn responses, conversations are managed by an autonomous Agent capable of reasoning, planning, and executing multi-step tasks by default</td></tr></tbody></table>


# Code Interpreter

This feature allows users to turn complex queries into actionable insights effortlessly by allowing the AI to run code, analyze data, and generate results in real time. This feature is accessible only through the Code Interpreter Agent.

<figure><img src="/files/D5I2cQwNuLs1Sir5GxP8" alt=""><figcaption></figcaption></figure>


# Browser Use

Browser Use allows the AI Agent to browse the web in real time to retrieve up-to-date information , verify facts beyond its trained knowledge and further interaction that can be done via browser

Browser Use is currently accessible through AI Agent mode, meaning it must be invoked within an Agent enabled conversation. The system automatically handles the search process, summarizes retrieved content, live browsing and replay browsing activity video

<figure><img src="/files/8jyNYV4Oz8KAQn6nKAa2" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/B8yrOlbMBep5IX96lbop" alt="" width="563"><figcaption></figcaption></figure>


# Knowledge Graph

Improves GLChat's comprehension by introducing structured relationships among various entities. This enhancement not only enhances retrieval accuracy and contextual reasoning but also sets the stage for the development of more insightful, connected, and contextually aware responses.\
\
By systematically organizing data, GLChat can identify and associate relevant information more efficiently, thereby improving the overall quality and intelligence of its interactions. This foundational improvement is pivotal in evolving GLChat into a more intuitive and sophisticated communication tool.


# Observability

## Observability in GLChat

### Introduction

Observability in GLChat is crucial for maintaining seamless communication and ensuring the platform's reliability and performance. It involves monitoring, alerting, and diagnosing issues to provide insights into application operations. This document explores the components and benefits of implementing robust observability practices in GLChat.

### Components of Observability

1. **Monitoring**: Continuously track system metrics such as latency, throughput, error rates, and resource usage.
2. **Logging**: Keep detailed logs of events and transactions occurring within the system for troubleshooting and auditing purposes.
3. **Tracing**: Follow the journey of a request through the system to understand dependencies, performance bottlenecks, and failures.

Incorporating these components in GLChat enhances our ability to detect and respond to issues swiftly.

### Benefits of Implementing Observability

Implementing observability within GLChat provides multiple advantages:

* **Enhanced User Experience**: By proactively identifying and resolving problems, users enjoy seamless and uninterrupted service.
* **Rapid Diagnoses**: Quickly pinpoint the root cause of system issues, reducing downtime and improving reliability.
* **Improved Performance**: Insights into system operations improve resource management and application performance.
* **Proactive Maintenance**: Detect trends and anomalies before they result in system failures.

The importance of observability in GLChat cannot be overstated. It ensures the platform operates smoothly, meeting user needs and business requirements. By leveraging an effective observability strategy, GLChat can maintain high standards of performance and user satisfaction, creating a robust communication tool for all its users.


# NLWeb

## Integration of GLChat with NLWeb

The integration of GLChat with NLWeb aims to elevate user interactions by combining advanced AI capabilities with web interfaces. This powerful synergy enables users to ask questions to the web in natural language, streamlining information retrieval and enhancing the overall user experience.

### Goals

* **Enhanced User Experience:** Provide a seamless and intuitive interface for users to interact with web resources.
* **Natural Language Processing (NLP):** Leverage NLP to allow users to ask questions in everyday language.
* **Advanced AI Capabilities:** Utilize the robust AI features of NLWeb to process and understand user queries efficiently.

### Benefits

1. **Intuitive Interactions:** Users can communicate with the web in natural language, similar to conversing with a human assistant.
2. **Efficiency in Information Retrieval:** Users acquire the information they seek more quickly and accurately without needing to use specific search terminologies.
3. **Advanced Understanding:** The integration leverages AI's deep learning capabilities to provide context-aware responses that meet user needs.

### Implementation Strategy

1. **Integration Setup:** Establish a connection between GLChat and NLWeb, ensuring seamless data flow and communication.
2. **NLP Optimization:** Enhance NLP capabilities to accurately interpret and respond to diverse user queries.
3. **User Feedback Mechanism:** Implement a feedback system for continual improvement based on user interactions and experiences.

### Future Outlook

The integration will continuously evolve, incorporating new AI advancements and user feedback to further bridge the gap between AI-powered assistance and user-friendly web interactions. This ongoing development aims to create a more dynamic and responsive user environment.


# Semantic Cache

## Semantic Cache in GLChat

GLChat, a cutting-edge chat application, employs an innovative feature called Semantic Cache. This feature is designed to enhance the efficiency and intelligence of the chat experience for users by optimizing the retrieval of relevant information.

### What is Semantic Cache?

Semantic Cache is a system that temporarily stores frequently accessed data based on its semantic meaning. This approach allows for quicker retrieval times and reduces the need for redundant data processing. By understanding the context and content of the chat, the cache can selectively store and manage data that is more likely to be reused.

### Benefits

1. **Improved Performance**: By keeping commonly required data readily available, the Semantic Cache significantly reduces response times. This results in faster information retrieval, creating a smoother user experience.
2. **Resource Efficiency**: Reducing redundant data retrieval saves computational resources, thereby minimizing unnecessary use of bandwidth and server load.
3. **Enhanced Personalization**: Semantic Cache enables GLChat to provide a more personalized experience, retaining information about user preferences and behaviors to tailor future interactions better.
4. **Scalability**: Efficient data management through caching allows GLChat to scale more effectively, handling more users and larger datasets without compromising performance.

### How It Works

1. **Data Analysis**: The system continuously analyzes chat data to identify patterns and frequently accessed information.
2. **Storage Strategy**: Once identified, this information is semantically stored in the cache, indexed by its contextual meaning rather than by simple keywords.
3. **Retrieval Optimization**: When a user requests information, GLChat looks into the Semantic Cache first, retrieving data that closely matches the semantic context of the request.
4. **Dynamic Updates**: The cache updates dynamically as conversations evolve, ensuring that the most relevant information is always prioritized.

The implementation of Semantic Cache in GLChat represents a significant step forward in optimizing chat applications. By understanding and anticipating user needs based on semantic understanding, GLChat delivers a more efficient, resource-friendly, and personalized chatting experience.

Adopting Semantic Cache technology ultimately makes GLChat not just a communication tool but an intelligent assistant in everyday interactions.


# Flexible Deployment

GLChat offers highly adaptable deployment options to suit various organizational needs and preferences. Whether you are looking to deploy on a Software as a Service (SaaS) model, within a Virtual Private Cloud (VPC), or on-premise, GLChat provides the flexibility and scalability required.

### SaaS Deployment

* **Ease of Use**: Enjoy hassle-free deployment with GLChat’s SaaS model, where all hosting and maintenance responsibilities are managed by us.
* **Scalability**: Seamlessly scale your chat capabilities as your user base grows, without worrying about infrastructure limitations.
* **Automatic Updates**: Benefit from regular software updates and feature enhancements without any manual intervention.

### Virtual Private Cloud (VPC) Deployment

* **Increased Security**: Deploy GLChat within your VPC to enhance security measures through dedicated network resources.
* **Customization**: Tailor your chat environment to meet specific organizational policies and compliance requirements.
* **Controlled Access**: Manage access and permissions through your cloud provider, ensuring sensitive data remains protected.

### On-Premise Deployment

* **Total Control**: Run GLChat on your own servers to maintain complete control over your infrastructure.
* **Data Sovereignty**: Ensure data remains within your defined geographical boundaries by adhering to data residency requirements.
* **Integration Flexibility**: Integrate GLChat with other on-premise systems and applications for seamless operational workflows.

Each deployment option is designed to meet diverse business needs, providing the flexibility to choose the right solution for your organization’s operational requirements and strategic goals.


# Anonymous Logging

GLChat offers a unique privacy-first feature known as anonymous logging. This innovative design ensures that user information is kept confidential by disguising user identities through advanced encryption methods. By prioritizing user privacy, GLChat allows individuals to communicate freely without fear of their data being exposed.

## Logs

Our logging system automatically detects and redacts sensitive information to maintain your privacy during system diagnostics.

**What is censored in logs?**

* KTP number
* NPWP number
* Family card number
* BPJS Number
* Phone number
* Email address
* International bank account number
* Credit card number
* Cryptocurrency wallet number
* IP address
* URL
  * Facebook account
  * LinkedIn account

## Traces

GLChat extends the same privacy-first approach to error tracking and traces monitoring. Sensitive user information from the captured traces and error will be protected.

**What is censored in traces?**

1. Credit card numbers
2. Password
3. API Key
4. Bearer token
5. Request cookies.
6. HTTP Message
7. Chat Message
8. Chat History
9. User ID
10. Query

## Our Commitment

With anonymous logging, users can rest assured that their conversations and personal data remain secure across all monitoring systems, reflecting GLChat's commitment to safeguarding user privacy above all else.


# GLChat Live

GLChat has evolved from a text-only assistant into a comprehensive multimodal conversational platform. It enables bidirectional streaming of inputs and outputs, thus expanding GLChat's ability to provide real-time, multimodal responses.

#### Example Use Case

1. Access advanced multimodal reasoning through the Live feature.
2. Support audio and video inputs.
3. Real-time responses with higher accuracy and context awareness.

#### Steps to Access the Live Feature in GLChat

* Open GLChat and select the chat where you want to use the Live feature.
* In the message input toolbar, tap the waveform icon to access Live.

<figure><img src="/files/n2cUVI6XVhbPC9B7Jc2t" alt=""><figcaption></figcaption></figure>

* On the Live page, click the microphone button to start capturing audio input, or click the camera button for video input.

<figure><img src="/files/j4NHJSdf3XJigiB1dV8d" alt=""><figcaption></figcaption></figure>

* To show the subtitles, user can click the caption button (second button from the right)

<figure><img src="/files/fVeJsK4lGdcQXditQgEl" alt=""><figcaption></figcaption></figure>

* Once you want to stop, click the stop button (x button on the far right) to end the recording. You will then be directed to the chat page.

<figure><img src="/files/4fayWV5b1tFSz8F2aC2T" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/futETJnRfdPRMWHmoJfa" alt=""><figcaption></figcaption></figure>


# CAPTCHA

Adds CAPTCHA verification during login to help prevent automated bot access

The system enhances security by significantly reducing the risk of abuse through actively monitoring and identifying suspicious login attempts. It diligently blocks automated login activities and unauthorized access attempts.\
\
By implementing advanced algorithms and machine learning techniques, the CAPTCHA system detects and responds promptly to unusual patterns and anomalies, ensuring that access remains secured and that legitimate users are protected from potential identity theft or fraudulent activity. This proactive approach not only shields against unauthorized entry but also provides peace of mind to users, ensuring a seamless and safe login experience.

<figure><img src="/files/mKGiZQ0x1IX57VrLMw9Y" alt=""><figcaption></figcaption></figure>


# Account Lockout

Temporarily locks an account after repeated failed login attempts

#### Feature: Account Lockout Mechanism

**Overview**

The account lockout mechanism is a security feature that temporarily locks an account after a series of consecutive failed login attempts. This control is designed to enhance security by preventing unauthorized access and mitigating brute-force attacks where malicious actors attempt to gain access to accounts by guessing passwords repeatedly.

**How It Works**

1. **Failed Login Attempt Tracking**: Every failed login attempt is logged and counted.
2. **Threshold Limit**: A predetermined number of failed attempts—commonly three to five—is set as the threshold.
3. **Temporary Lockout**: Once the threshold is reached, the account is temporarily locked, preventing further login attempts.

**Benefits**

* **Security Enhancement**: By limiting the number of login attempts, the lockout mechanism significantly reduces the risk of unauthorized access due to credential guessing or automated attacks.
* **Account Protection**: Users are alerted to suspicious activities on their accounts, promoting vigilance and encouraging password updates or two-factor authentication (2FA).
* **Operational Safeguards**: Automated account lockouts help maintain system integrity by ensuring that only legitimate users can access their accounts while providing system administrators with alerts about potential security threats.

By implementing a robust account lockout mechanism, organizations can proactively safeguard their systems and users against unauthorized intrusions while maintaining user trust and data privacy.

<br>

<figure><img src="/files/I581pVGqfpA54lHespBe" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/BcmyKfoBWeqfLyNHBZRm" alt=""><figcaption></figcaption></figure>


# Multi-Factor Authentication

Adds a second verification step during login to strengthen account security.

### Two-Factor Authentication: Enhancing Account Security

In today's digital landscape, protecting user accounts is more critical than ever. A single password is often inadequate to safeguard accounts from unauthorized access, particularly with the increasing sophistication of cyber threats. To address this vulnerability, implementing a second verification step during the login process has become essential.

#### Benefits of Two-Factor Authentication

1. **Increased Security**: By requiring a second factor of authentication, user accounts gain an additional layer of defense. Even if hackers manage to obtain a user's password, the secondary verification step serves as a barrier to unauthorized entry.
2. **Mitigation of Password Compromises**: With two-factor authentication, a compromised password alone is insufficient for access. Users must also provide a unique second factor, such as a one-time code sent to their mobile device or biometrics like a fingerprint.
3. **Peace of Mind**: Knowing that their accounts are fortified with an extra security measure, users experience greater peace of mind. Two-factor authentication assures users that their sensitive information and personal data are considerably more protected.
4. **Industry Compliance**: Many regulatory standards and industry best practices recommend or require the use of two-factor authentication. By adopting this security measure, organizations can ensure compliance with these guidelines and protect themselves legally.

Note: Multi-factor authentication is configured via the admin dashboard by your organization's admin.<br>

<figure><img src="/files/gMKZKH02i6DbbzArK1sP" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/gbFGoGmwVOHmzZRDlx4W" alt=""><figcaption></figcaption></figure>


# Seamless Sign-In with Identity Provider SSO

GLChat supports Identity Provider Single Sign-On (SSO) integration so users can access GLChat using the same account they already use in your organization’s application.

This is useful when GLChat is embedded as a widget or integrated into an existing client platform. Instead of asking users to log in again, GLChat can recognize the authenticated user from the client system and provide a smoother access experience.

### What This Feature Does

With Identity Provider SSO integration, your organization can connect its existing login and authentication system with GLChat.

When a user opens GLChat from your application:

1. The user logs in to your application as usual.
2. GLChat receives the authenticated user identity from your system.
3. GLChat automatically signs in the user based on that identity.
4. The user can start using GLChat without a separate GLChat login step.

This helps create a seamless experience between your application and GLChat.

### When to Use This

This feature is recommended when:

* Your organization already has its own login page or authentication system.
* GLChat is embedded inside your website, portal, dashboard, or internal application.
* You want users to access GLChat without signing in separately.
* You want to reduce friction when introducing GLChat to existing users.

### User Experience

For new users, GLChat can automatically create or recognize the user based on the identity provided by your system.

For example, if a user is already logged in to your application, they can open the GLChat widget and immediately start chatting without entering their email or password again.

### Existing GLChat Account Handling

If the user already has an existing GLChat account with the same identifier, such as the same email address, GLChat will ask the user how they want to continue.

The user may choose to:

* **Merge or link the account** so their existing GLChat account can be connected with the company login.
* **Continue as a separate account** if they do not want to merge the accounts.

This helps avoid duplicate accounts while still giving users control over how their identity is handled.

### Example Scenario

<figure><img src="/files/vcLkNHaCig9wYT5xjwmI" alt=""><figcaption></figcaption></figure>

A company has an internal employee portal. Employees already sign in to the portal using the company’s authentication system.

When GLChat is embedded into the portal, employees do not need to log in to GLChat separately. After they sign in to the portal, they can open GLChat and start using it directly.

<figure><img src="/files/Iief7QiSQE4LNuwoVwSg" alt=""><figcaption></figcaption></figure>

If an employee already has a GLChat account using the same email, GLChat will ask whether they want to merge the account or continue separately.

<figure><img src="/files/cUHeKmG4ateX5dM70q1S" alt=""><figcaption></figcaption></figure>

### Benefits

Identity Provider SSO integration helps organizations:

* Provide a smoother login experience for users.
* Reduce repeated login steps.
* Keep GLChat access aligned with the organization’s existing authentication system.
* Support embedded GLChat use cases inside client-owned platforms.
* Reduce confusion caused by multiple accounts or separate login flows.

### Notes

The exact integration flow may depend on your organization’s authentication setup. \
For integration, refer to the page below:

[API Credentials](/glchat/organization-admin/how-to-guides/integration-and-services/api-credentials#idp-sso-integration-key)

[Partner SSO](/glchat/developer-documentation/partner-sso)


# Sign Out

To sign out, simply click your profile menu in the bottom-left corner and select the **Sign Out** option.

{% tabs %}
{% tab title="Website" %}

<figure><img src="/files/lr54EU1UIdOocWw3AkMO" alt=""><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Mobile (iOS/Android)" %}
![](/files/IfvSU0lR9wj1uIESCd25)
{% endtab %}
{% endtabs %}


# Resources

{% content-ref url="/pages/VcjAWvC9LFTrVJ7ks6VA" %}
[Acronyms and Terminologies](/glchat/organization-admin/resources/acronyms-and-terminologies)
{% endcontent-ref %}


# Acronyms and Terminologies

This section provides definitions and explanations for key terms and acronyms used throughout the GLChat to help users, both technical and non-technical, to understand the platform’s components, features, and configuration settings more easily. Refer to this page whenever you encounter unfamiliar terminology in the documentation.

| Term/Acronym                                       | Description                                                                                                                                                                                                              |
| -------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| A2A (Agent-to-Agent)                               | A system where one AI agent can work together with another, sharing tasks or passing information to complete complex actions more effectively.                                                                           |
| API (Application Programming Interface)            | A way for different software systems to talk to each other and exchange information or functions. It helps apps work together behind the scenes.                                                                         |
| API Key                                            | A unique code used to securely connect one system or app to another. It works like a password for software, allowing access while keeping things protected.                                                              |
| Augment                                            | Enhancing the system's ability by adding extra information or support from other tools or data.                                                                                                                          |
| Batch Size                                         | The number of items processed by the system at once, like documents or data chunks.                                                                                                                                      |
| Bearer Token                                       | A secure access key used to verify a user's identity when making system requests.                                                                                                                                        |
| Blacklist                                          | A list of blocked or restricted websites that are given special permission to access something.                                                                                                                          |
| Cache                                              | Temporary storage that helps speed up repeated tasks or data retrieval.                                                                                                                                                  |
| Char (Character)                                   | Each letter, number, space, or symbol in a text.                                                                                                                                                                         |
| Chunker                                            | A tool that breaks long text into smaller pieces (or “chunks”) to help the AI process the content more effectively and stay within its limits.                                                                           |
| Docproc (Document Processing)                      | Systems that read, understand, and extract data from documents.                                                                                                                                                          |
| DPO (Document Processing Orchestrator)             | A process that helps prepare and organize documents so they can be used as reliable sources of information in the system.                                                                                                |
| Email Domain                                       | The part of an email address that comes after the “@” symbol (e.g., in <user@example.com>, the domain is example.com).                                                                                                   |
| Formatter                                          | A tool that changes how information is displayed or structured, so it looks cleaner or follows a standard.                                                                                                               |
| Graph RAG Indexer                                  | A smart indexing tool that uses a graph (network-like structure) to help AI find and retrieve better answers.                                                                                                            |
| HTTP (HyperText Transfer Protocol)                 | The standard way data is sent and received over the web.                                                                                                                                                                 |
| Index Name                                         | The name of the vector index where the knowledge base content is stored. It’s auto-generated by default but can be customized if needed.                                                                                 |
| Knowledge Base                                     | A collection of information or documents that the chatbot can refer to when answering questions.                                                                                                                         |
| Kwargs (Keyword Arguments)                         | Extra options or settings you can include when using a feature, by naming them clearly. It helps the system understand exactly what you want to change or customize.                                                     |
| LLM (Large Language Model)                         | A type of Language Model that has been trained with a huge amount of text data, making it much more powerful and capable. It can handle complex instructions, answer detailed questions, and hold natural conversations. |
| LLM Labs                                           | A platform used to test, manage, and connect AI models (like LLMs) to other parts of the system, such as document storage or search tools.                                                                               |
| LM (Language Model)                                | An AI model that can understand and generate text, like answering questions or continuing a sentence. It can be small or large depending on how much it has been trained.                                                |
| LM Invoker                                         | A component that takes your prompt (question or instruction) and sends it to the AI model (LM), then returns the response.                                                                                               |
| LMRP (Language Model Request Processor)            | A module that helps the AI prepare your question, send it to the language model, and clean up the answer before showing it to you. It ensures the whole process, from asking to answering, is smooth and accurate.       |
| MCP (Model Context Protocol)                       | A system that allows the chatbot to use extra tools, like custom code or functions, to perform more advanced tasks.                                                                                                      |
| Model                                              | In this context, model refers to a language model, the part of the AI that has learned to understand and respond to human language based on large amounts of training data.                                              |
| Multimodal                                         | A system that can understand and process more than one type of input, such as text, images, or audio.                                                                                                                    |
| OAuth (Open Authorization)                         | A secure way to connect one app to another without sharing your password. For example, an admin can approve access through a permission screen to link GLChat with services (e.g. Google Workspace).                     |
| PII (Personally Identifiable Information)          | Sensitive data like names, emails, phone numbers, etc., which can be anonymized.                                                                                                                                         |
| PII Anonymization                                  | The process of hiding or replacing personal information (like names, phone numbers, or ID numbers) to protect user privacy before the data is processed or stored.                                                       |
| Pipeline                                           | A step-by-step process the system follows to handle a chat or a document, from receiving input to giving a response.                                                                                                     |
| Preset                                             | A saved configuration that controls how the system behaves, including settings like supported file types, privacy rules, and search options.                                                                             |
| Prompt                                             | The message or instruction you type to the chatbot, or the internal guide the system uses to decide how to respond.                                                                                                      |
| RAG (Retrieval-Augmented Generation)               | A method that enhances LLM responses with relevant external information from a knowledge base.                                                                                                                           |
| RAGO (Retrieval-Augmented Generation Orchestrator) | A GLChat pipeline type used to configure how information is retrieved and presented during chat interactions.                                                                                                            |
| Rerank                                             | Re-sorting search results to show the most relevant ones at the top.                                                                                                                                                     |
| SSE (Server-Sent Events)                           | A way for the system to send real-time updates from the server to the user.                                                                                                                                              |
| SSO (Single-Sign On)                               | A login method that lets users access multiple systems using one account, usually their work email.                                                                                                                      |
| Threshold                                          | A set limit used by the system to decide whether something is accepted or rejected.                                                                                                                                      |
| Timeout                                            | The maximum time allowed for an AI or tool to respond. If it takes too long, the system will stop waiting and cancel the request.                                                                                        |
| Transport Type                                     | The method used to deliver responses from the system to the user (e.g., HTTP or SSE).                                                                                                                                    |
| Vector Store                                       | A type of database where document meanings (not just words) are stored in a way the AI can search by understanding the context. It helps the system find the most relevant answers quickly.                              |
| Vector Weight                                      | A value showing how important a piece of information is when comparing or searching using AI.                                                                                                                            |
| Vectorizer                                         | A tool that converts text into a numeric format (called vectors) so that the AI can understand and work with the content mathematically.                                                                                 |
| Whitelist                                          | A list of allowed websites that are given special permission to access something.                                                                                                                                        |


# FAQ

❓**Q: Which browsers support GLChat?**\
💡A : GLChat works on all modern, up-to-date web browsers that follow current web standards. We recommend using the latest versions of:

1. Google Chrome
2. Microsoft Edge
3. Mozilla Firefox
4. Safari (macOS & iPadOS)

For the best experience, please ensure:

1. Your browser is updated within the last few years
2. JavaScript and standard browser features are enabled

⚠️ Note: Very old browsers or outdated mobile browsers may not fully support some features.

❓**Q:** What does “modern browser” mean technically?

💡A : GLChat follows widely adopted web standards introduced around 2022, which are now supported by most mainstream browsers. You can view the reference list of supported browsers [here](https://browsersl.ist/#q=baseline+2022%0Anot+iOS+%3C+16.4%0A)

❓**Q: I encountered error "Invalid regular expression: invalid group specifier name" when using Safari browser.**\
💡A : Update your Safari to version 16.4 or later on macOS or iOS to resolve issues with earlier versions, as GLChat supports Safari 16.4 onwards.


# Features - Work in Progress

Last update: March 13, 2026

Below are the list of features that currently are work in progress:

<table><thead><tr><th width="74">No</th><th>Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td>Image Processing - Image Understanding</td><td>Capability to process image from context in a document or website so that it will enrich the answer</td></tr><tr><td>2</td><td>Canvas</td><td>Collaborate on rich content inside the chat</td></tr><tr><td>3</td><td>Long Term Memory Implementation</td><td>Retain key user info across sessions</td></tr><tr><td>4</td><td>Deep Think</td><td>Enable deeper, multi-turn reasoning and analysis</td></tr><tr><td>5</td><td>Progressive Web Apps (PWA)</td><td>Install GLChat as a Progressive Web App for seamless access</td></tr><tr><td>6</td><td>Computer use</td><td>Enable GLChat to perform desktop tasks like a human assistant—clicking, typing, and navigating applications</td></tr><tr><td>7</td><td>Enhancement Auto FAQ Cache for GLChat</td><td>Enhancement Auto FAQ from knowledge base focused to more on application focused, including the experience of creating Auto FAQ.</td></tr><tr><td>8</td><td>Improvement Notification of Asynchronous Task in GLChat application</td><td>Delivers real-time in-app and email alerts when long-running operations (e.g., Deep Research, Agents) complete or encounter errors.</td></tr><tr><td>9</td><td>Pin Chat / Star Chat</td><td>Enables users to mark important conversations for quick access, with a dedicated “Pinned Chats” view in the sidebar.</td></tr><tr><td>10</td><td>Evaluator in GLChat</td><td>Integrates configurable evaluation pipelines—combining multi-LLM scoring, automated metrics, and optional human-in-the-loop review—to assess response relevance, accuracy, and compliance, with a dashboard for tracking performance over time</td></tr><tr><td>11</td><td>1 Project with Multiple Knowledge Bases</td><td>Learn how to streamline your workflow by integrating multiple knowledge bases within a single project. This feature allows you to consolidate information, improving accessibility and efficiency in managing project resources.</td></tr><tr><td>12</td><td>Auto / Smart Routing Pipeline</td><td><p>This system directs user queries to the best response method:</p><ul><li><strong>Chat with LLM</strong>: Offers interactive, real-time conversations.</li><li><strong>AI Agent</strong>: Executes tasks automatically with AI logic.</li><li><strong>Search</strong>: Provides quick, relevant information retrieval.</li></ul></td></tr></tbody></table>


# Sitemap

Sitemap of Project User Interface

<figure><img src="/files/alGB8B5UygLrhL8lryrV" alt=""><figcaption></figcaption></figure>

Mindmap of Project User Interface

<figure><img src="/files/yncZMuNVe73iEXW0glF4" alt=""><figcaption><p>Screenshot of Application UI Mindmap</p></figcaption></figure>


# Overview

The **Organization Admin Dashboard** acts as the operational backbone of **GLChat**, providing administrators with complete visibility and control over the platform’s configuration and usage. Each module within the dashboard is designed to **simplify setup**, **streamline management**, and **accelerate configuration workflows**, ensuring that both technical and non-technical users can easily tailor GLChat to their organizational needs.


# Introduction

The **Organization Admin Dashboard** is the central control panel within **GLChat Settings**, enabling administrators to manage configurations and customize the platform to meet organizational requirements. It serves as the primary interface for setting up **agents**, **tools**, **knowledge bases**, and other **access control settings**—all in one unified place.

#### 💡 What You Can Do

Through the Admin Dashboard, you can:

1. **Configure applications**
2. **Manage users, roles, and permissions** for secure access
3. **Set up AI Agents** and integrate external **tools or APIs**
4. **Manage pipelines, presets, and other configurations** for enhanced workflow control
5. and many more..

This section provides an overview of each component within the Admin Dashboard, explaining how configurations integrate and interact within the broader **GLChat ecosystem**.

For detailed step-by-step instructions, please refer to the [**How-To Section**](/glchat/sdk/how-to-guides).


# Features Overview

## General

<table><thead><tr><th width="70.24380493164062">No</th><th width="269.63427734375">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td>Login</td><td><p>The application's ability to display a login page with various sign-in options. Currently supports:</p><ol><li>Basic authentication (username and password)</li><li>Sign in with Google</li><li>Sign in with Azure Entra ID</li><li>Sign in with GitHub</li></ol></td></tr></tbody></table>

## GLChat Project

### Analytics

<table><thead><tr><th width="70.24380493164062">No</th><th width="269.63427734375">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/1yhEap1v5cra7kqp3YPM">/pages/1yhEap1v5cra7kqp3YPM</a></td><td>Monitor GLChats <strong>user activity, engagement, and performance metrics</strong> using the Analytics Dashboard.</td></tr></tbody></table>

### Project

<table><thead><tr><th width="69.95379638671875">No</th><th width="269.80718994140625">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/XvL3ZuxZSZ2u97zsInoz">/pages/XvL3ZuxZSZ2u97zsInoz</a></td><td>Create, view, edit, and delete GLChat projects and manage their configurations.</td></tr><tr><td>2</td><td><a data-mention href="/pages/xapTXVQsUFdl3pZZRdPi">/pages/xapTXVQsUFdl3pZZRdPi</a></td><td>Build and manage a curated list of frequently asked questions that appear as suggested prompts in the project.</td></tr><tr><td>3</td><td><a data-mention href="/pages/R4AO7awm0I18lk2c99wF">/pages/R4AO7awm0I18lk2c99wF</a></td><td>Create and manage reusable prompt templates that guide how the project interacts with AI models.</td></tr><tr><td>4</td><td><a data-mention href="/pages/965mjb8QiL2I7x8MJ29K">/pages/965mjb8QiL2I7x8MJ29K</a></td><td>Manage system-level prompt templates that control how AI responses are generated and structured.</td></tr><tr><td>5</td><td><a data-mention href="/pages/e1Qceqk363MmwOPoQt2n">/pages/e1Qceqk363MmwOPoQt2n</a></td><td>Configure and customize the GLChat widget to match your user needs and channel requirements.</td></tr></tbody></table>

### Knowledge Base

<table><thead><tr><th width="70.1363525390625">No</th><th width="269.73992919921875">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/B6XS9H56Uo2ACHOZyFh1">/pages/B6XS9H56Uo2ACHOZyFh1</a></td><td>Set up the LLM Labs Client to enable integration with GLChat.</td></tr><tr><td>2</td><td><a data-mention href="/pages/bH75yYpbTtPIA1CokKZU">/pages/bH75yYpbTtPIA1CokKZU</a></td><td>Upload, organize, and maintain a centralized repository of knowledge used to generate accurate AI responses.</td></tr><tr><td>3</td><td><a data-mention href="/pages/08tTCHpO3GDWliD1ArUu">/pages/08tTCHpO3GDWliD1ArUu</a></td><td>Create and manage FAQs from your Knowledge Base. These FAQs appear in the ‘<a href="/pages/xapTXVQsUFdl3pZZRdPi">Question Library</a>’ as suggested prompts for users.</td></tr><tr><td>4</td><td><a data-mention href="/pages/CtJeaSfGjQdWaDV7UjJk">/pages/CtJeaSfGjQdWaDV7UjJk</a></td><td>Connect and manage external sources (such as SharePoint) to sync third-party content into your Knowledge Base.</td></tr></tbody></table>

#### AI Agent

<table><thead><tr><th width="70.18121337890625">No</th><th width="269.7979736328125">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/vZN2YCTUzOrOLiGhWJ4L">/pages/vZN2YCTUzOrOLiGhWJ4L</a></td><td>Create and manage AI agents (built-in or Langflow-based) with customizable behaviors and workflows.</td></tr><tr><td>2</td><td><a data-mention href="/pages/f7WpLtRApAFPTnGuXZEC">/pages/f7WpLtRApAFPTnGuXZEC</a></td><td>Add and manage tools that can be integrated into an agent’s workflow to extend its functionality.</td></tr></tbody></table>

## Developer Mode

### Pipeline and Preset

<table><thead><tr><th width="69.9346923828125">No</th><th width="270.42138671875">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/w7NPVXj75lkID60wnjM9">/pages/w7NPVXj75lkID60wnjM9</a></td><td>Set up and manage Retrieval-Augmented Generation (RAG) pipelines that control how data is retrieved and used in AI responses.</td></tr><tr><td>2</td><td><a data-mention href="/pages/nIDzYYDnB7CE3yB9Nbuj">/pages/nIDzYYDnB7CE3yB9Nbuj</a></td><td>ConConfigure Document Processing Orchestrator (DPO) pipelines to manage how documents are processed and integrated with the Knowledge Base.</td></tr><tr><td>3</td><td><a data-mention href="/pages/EzP6XXPmeQYz0z1uBsWM">/pages/EzP6XXPmeQYz0z1uBsWM</a></td><td>Use schema-based pipelines to define structured document processing workflows.</td></tr><tr><td>4</td><td><a data-mention href="/pages/9vVVq1GVTbRgyPiV9saS">/pages/9vVVq1GVTbRgyPiV9saS</a></td><td>Manage database configurations for the Knowledge Store used by the system.</td></tr></tbody></table>

### LLM Model

<table><thead><tr><th width="70.2962646484375">No</th><th width="269.73614501953125">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/WyEqcOu2emYS4iKwuJPQ">/pages/WyEqcOu2emYS4iKwuJPQ</a></td><td>Register and manage external LLM providers by adding API credentials, enabling access to models from multiple sources.</td></tr><tr><td>2</td><td><a data-mention href="/pages/M4wC2XwQ4jOHyWN3BceI">/pages/M4wC2XwQ4jOHyWN3BceI</a></td><td>Register, configure, and manage the AI models used within the system.</td></tr></tbody></table>

## Organization

### User & Team Management

<table><thead><tr><th width="69.65972900390625">No</th><th width="270.35882568359375">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/iB10QW9NI5UlYSbQJ7Ta">/pages/iB10QW9NI5UlYSbQJ7Ta</a></td><td>Create and manage user accounts, assign roles and permissions, and control access across the platform.</td></tr><tr><td>2</td><td><a data-mention href="/pages/THWAYbOSJp0m07uZj2QN">/pages/THWAYbOSJp0m07uZj2QN</a></td><td>CreaCreate and manage teams, and assign them access to applications.</td></tr></tbody></table>

### Organization Management

<table><thead><tr><th width="70.21710205078125">No</th><th width="270.01416015625">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/ZJoTmmaPRzg9VsCPvTxc">/pages/ZJoTmmaPRzg9VsCPvTxc</a></td><td>Manage your organization’s profile, including name, branding, and general settings.</td></tr><tr><td>2</td><td><a data-mention href="/pages/IjcAXizjBcLW37oUKygL">/pages/IjcAXizjBcLW37oUKygL</a></td><td>View and monitor subscription information, including plan details, usage metrics, and renewal status.</td></tr></tbody></table>

### Integration and Services

<table><thead><tr><th width="70.21710205078125">No</th><th width="270.01416015625">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/DdBCyUtTy8iVKr6NsdR2">/pages/DdBCyUtTy8iVKr6NsdR2</a></td><td>Create and manage API keys for external LLM providers and integrations.</td></tr><tr><td>2</td><td><a data-mention href="/pages/ld81hhOAHy4iIcSLhOfY">/pages/ld81hhOAHy4iIcSLhOfY</a></td><td>Connect or disconnect supported third-party services (such as Google Drive, Slack, and GitHub) to integrate external tools into the platform.</td></tr><tr><td>3</td><td><a data-mention href="/pages/Sqzl7Tq5xMG4zT3BScHv">/pages/Sqzl7Tq5xMG4zT3BScHv</a></td><td>Configure third-party communication channels (such as WhatsApp) to enable messaging integration with GLChat applications.</td></tr></tbody></table>

## Audit Log

<table><thead><tr><th width="70.4549560546875">No</th><th width="270.03192138671875">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/H9IzuBNWFEWyBAramyYV">/pages/H9IzuBNWFEWyBAramyYV</a></td><td>Record and review all approval activities performed through the Human-in-the-Loop (HITL) process.</td></tr></tbody></table>

## Guardrails

<table><thead><tr><th width="70.4549560546875">No</th><th width="270.03192138671875">Features</th><th>Description</th></tr></thead><tbody><tr><td>1</td><td><a data-mention href="/pages/2yxc8PfJ0FVQ85452d7E">/pages/2yxc8PfJ0FVQ85452d7E</a></td><td>Define and enforce safety and behavior rules by validating inputs and/or outputs to ensure compliance, relevance, and appropriateness.</td></tr></tbody></table>


# Getting Started

After logging in, you will see the **Project** view by default. To switch to the Admin Dashboard:

1. Click on your profile icon on the bottom left sidebar.
2. Select **‘Switch to Admin Dashboard’** from the dropdown menu.

<figure><img src="/files/wYoezz2vJ9Q12MJE8R4N" alt=""><figcaption></figcaption></figure>

After clicking, you will be switched to the Organization Admin Dashboard page.


# How-To Guides

This guide offers a step-by-step walkthrough of GLChat’s back office (Admin Dashboard) capabilities to help get the most out of the platform and provides a quick overview of what it can do. It focuses on practical use of each feature, with clear instructions, screenshots, and tips for both everyday tasks and advanced use cases.

{% content-ref url="/pages/1yhEap1v5cra7kqp3YPM" %}
[Analytics](/glchat/organization-admin/how-to-guides/analytics)
{% endcontent-ref %}

{% content-ref url="/pages/XvL3ZuxZSZ2u97zsInoz" %}
[Project](/glchat/organization-admin/how-to-guides/project)
{% endcontent-ref %}

{% content-ref url="/pages/RuVEHwITDWKfXvr4hgZV" %}
[Knowledge Base](/glchat/organization-admin/how-to-guides/knowledge-base)
{% endcontent-ref %}

{% content-ref url="/pages/vZN2YCTUzOrOLiGhWJ4L" %}
[AI Agent](/glchat/organization-admin/how-to-guides/ai-agent)
{% endcontent-ref %}

{% content-ref url="/pages/Mf5C0w4QLWFZPqAKyv6H" %}
[Pipeline and Preset](/glchat/organization-admin/how-to-guides/pipeline-and-preset)
{% endcontent-ref %}

{% content-ref url="/pages/M4wC2XwQ4jOHyWN3BceI" %}
[LLM Model](/glchat/organization-admin/how-to-guides/llm-model)
{% endcontent-ref %}

{% content-ref url="/pages/iB10QW9NI5UlYSbQJ7Ta" %}
[User Management](/glchat/organization-admin/how-to-guides/user-management)
{% endcontent-ref %}

{% content-ref url="/pages/THWAYbOSJp0m07uZj2QN" %}
[Team Management](/glchat/organization-admin/how-to-guides/team-management)
{% endcontent-ref %}

{% content-ref url="/pages/ZJoTmmaPRzg9VsCPvTxc" %}
[Organization Management](/glchat/organization-admin/how-to-guides/organization-management)
{% endcontent-ref %}

{% content-ref url="/pages/QDSWUKScbyLJRKn092eY" %}
[Audit Log](/glchat/organization-admin/how-to-guides/audit-log)
{% endcontent-ref %}

{% content-ref url="/pages/2yxc8PfJ0FVQ85452d7E" %}
[Guardrail](/glchat/organization-admin/how-to-guides/guardrail)
{% endcontent-ref %}


# Analytics

To access this menu, go to the sidebar and select Analytics.

This menu allows users to monitor GLChat's **user activity, engagement, and performance metrics** using the Analytics Dashboard.

<figure><img src="/files/PvEmzQ8McrW5PLA4Ws1Z" alt=""><figcaption></figcaption></figure>

***

## Analytics Dashboard Overview

The Analytics Dashboard consists of the following sections:

***

### 1. User Metrics Overview

<figure><img src="/files/XQUYPuJZLG4wZg9LUstt" alt=""><figcaption></figcaption></figure>

Provides a summary of user activity and seat usage.

* **Total Guests** – Users who accessed without logging in
* **Total Logged In Users** – Users who signed into the platform
* **Total Chat Users** – Users who actively used chat
* **Logged In Users (Utilization)** – % of seats used by logged-in users
* **Chat Active Users (Utilization)** – % of seats used by active chat users

### 2. Conversation Metrics

<figure><img src="/files/GNB3q84QCkYzZcSJ7SHt" alt=""><figcaption></figcaption></figure>

Measures platform usage and communication activity.

* **Total Conversations** – Number of conversations
* **Total Messages** – Number of messages sent
* **Average Response Time** – Time taken to respond
* **Average Messages per Conversation** – Depth of conversations

### 3. Charts & Trends

<figure><img src="/files/RTmaTE1cFMUQeR35Rjb8" alt=""><figcaption></figcaption></figure>

Visual representation of usage patterns.

* **Overview Trends** – Users, messages, conversations over time
* **Conversation Trends** – New vs active conversations
* **Conversations by Hour** – Activity distribution across the day
* **User Activity** – New, returning, and guest users

### 4. User Activity Table

<figure><img src="/files/j2Ys1bEjnt901E3QlAGU" alt=""><figcaption></figcaption></figure>

The User Activity Table provides a detailed breakdown of each user’s activity within the platform.

* **User ID** — Unique identifier assigned to the user
* **User Name** — Name of the user
* **Last Active** — Date and time of the user’s most recent activity
* **Total Conversations** — Total number of conversations created by the user
* **User Activity Details** — Click the **View** button to see detailed activity information for the selected user

<figure><img src="/files/ODYG4wy22J6TQK5C9nbk" alt=""><figcaption></figcaption></figure>

* **Export User Activity** — Click the **Export** button to download user activity data for reporting or analysis purposes.

***

## Select Time Range

Use the **time filter (top-right corner)** to adjust your data view. Changing the time range updates *all metrics and charts* across the dashboard.

**Available options:**

* Today
* 1 day ago
* 1 week ago
* 1 month ago
* Custom Range (up until a month)

***


# Usages

To access this menu, go to the sidebar and select Subscription > Plan.

<figure><img src="/files/QXhHpW4sg1HSa8yw8UUl" alt=""><figcaption></figcaption></figure>

The **Usages** menu allows users to view the current usage of their organization’s subscription plan. This page provides an overview of quotas, usage limits, and subscription-related information.

### Organization Summary

At the top of the page, users can view the organization’s subscription details, including:

| Field              | Description                                    |
| ------------------ | ---------------------------------------------- |
| Organization ID    | Unique ID of the organization                  |
| Organization Name  | Name of the registered organization            |
| Account Created    | Date when the organization account was created |
| License Expiration | Date when the subscription license will expire |

This section helps users confirm which organization they are managing and check the validity period of their subscription.

### Usage Overview

The **Usages** section displays the current quota usage for the organization.

Users can monitor the following items:

| Usage Type | Description                                                        |
| ---------- | ------------------------------------------------------------------ |
| Admin      | Number of admin accounts currently used out of the available quota |
| Projects   | Number of projects created out of the available quota              |
| Message    | Number of messages used out of the available monthly quota         |
| User       | Number of user accounts currently used out of the available quota  |

Each usage card shows:

* The current usage count
* The maximum quota allowed by the subscription plan
* A progress bar to visually indicate usage consumption

### Message Quota Per User

The **Message Quota Per User** table shows the monthly message quota assigned to each user.

The table includes:

| Column                  | Description                                                    |
| ----------------------- | -------------------------------------------------------------- |
| Username                | Email address or username of the user                          |
| Message Quota per Month | Number of messages used by the user out of their monthly quota |

{% hint style="info" %}
Usage data is based on the organization’s active subscription plan. If the usage reaches the maximum quota, users may need to reduce usage or upgrade their subscription plan.
{% endhint %}

\ <br>


# Plan

In the ‘Plan’ menu, you can define and adjust the resource quotas available for the organization’s subscription.

***

## Resources

For the organization listed, you can configure the quota of:

1. Message: Set the total number of message credits available under the subscription plan.
2. Applications: Define how many chatbot applications can be created within the organization.
3. Admin: Specify the maximum number of administrator accounts allowed.

After adjusting the desired quotas, click the ‘Save’ button to apply the changes.

<figure><img src="/files/gSm6Sh80Mf3akP7HRBY5" alt=""><figcaption></figcaption></figure>

A confirmation message will appear once the plan has been updated.

<figure><img src="/files/gCwkSyibV2B1ZkwFoS1Y" alt=""><figcaption></figcaption></figure>

### **Message Quote Limit Reached**

This feature monitors message usage under the organization’s subscription plan.\
When the assigned message limit is exceeded, the system automatically restricts further message generation and notifies administrators.

<figure><img src="/files/c8GhG5gT6LPQGV1s3BnM" alt=""><figcaption></figcaption></figure>

While in the **Application UI**, if users attempt to send a new message after the limit is reached, the system will display an error message: **“Monthly message quota exceeded.”**

<figure><img src="/files/R4WpK0EAMrzkbZDUtrA0" alt=""><figcaption></figcaption></figure>

**Application Limit Reached**\
This feature enforces the application creation limit defined in the subscription plan. When the limit is reached, the **“New Application”** button in the **Application** section becomes inactive, preventing administrators from creating additional applications.

<figure><img src="/files/jlOf2QILQV62jF3VQoN4" alt=""><figcaption></figcaption></figure>

**Admin Limit Reached**

This feature enforces the maximum number of users allowed under the organization’s subscription plan. When administrators attempt to add a new user under any role after the limit is reached, the system will display an error message: **“Admin quota limit exceeded. Current: \[Current Quota]/\[Total Limit]”**

<figure><img src="/files/eF8j0QWPOBnzAGdPKIyS" alt=""><figcaption></figcaption></figure>

This indicates that no additional users can be added unless the subscription plan is upgraded or existing users are removed.


# Project

To access the Project menu, go to the sidebar and select Project

<figure><img src="/files/TQVnS120BnkAqknUMIGN" alt=""><figcaption></figcaption></figure>

A **Project** represents a specific use case or business solution within GLChat. Each project can contain one or more Knowledge Bases, Agents, prompts, configurations, and other supporting features required to deliver the intended user experience.

This menu allows administrators to manage all projects within GLChat. Users can create, update, and delete projects as needed.

## Create New Project

<figure><img src="/files/iZg5IeyRRVZR22UU39nW" alt=""><figcaption></figcaption></figure>

To create a new project, click **New Project** and follow the setup wizard.

1. Step 1: [Setup Project](/glchat/organization-admin/how-to-guides/project/setup-project)\
   Configure the basic project information and AI settings
2. Step 2: [Manage Access](/glchat/organization-admin/how-to-guides/project/manage-access)\
   Define who can access the project and assign the appropriate permissions.
3. Step 3: [Manage FAQ](/glchat/organization-admin/how-to-guides/project/manage-faq) (Optional)\
   Create frequently asked questions to help users quickly understand the project and discover common use cases.
4. Step 4: Configure [Channel](/glchat/organization-admin/how-to-guides/project/channel) (Optional)\
   Select the channels where the project will be available.

Once all required steps are completed, the project is ready for use.

***

## View and Edit Project Detail

To edit an existing project, click the **Edit** icon on the Project List page. You can modify project settings, access permissions, and channel options.

<figure><img src="/files/0nCC4HX4VCVEqc26pJcx" alt=""><figcaption></figcaption></figure>

When finished, click **Save** to apply your changes.

***

## Search Project

Use the **Search** field to find an project by typing its name.

<figure><img src="/files/spEBGeaGZaeQydOYsRka" alt="" width="449"><figcaption></figcaption></figure>

***

## Delete Project

To delete an project, click the **Trash** icon in the Action column. A confirmation pop-up will appear, click **Delete** to proceed or **Cancel** to abort.

<figure><img src="/files/IYX6vXPcZQfHggAvn39t" alt="" width="563"><figcaption></figcaption></figure>

{% hint style="info" %}
Note: You cannot delete an application that still has assigned users.\
Remove all assigned users first before attempting deletion.
{% endhint %}


# Setup Project

<figure><img src="/files/qNckFrivrur8LK6QyRCc" alt=""><figcaption></figcaption></figure>

Set up your GLChat project by configuring AI models, Knowledge Bases, agents, user access, and other project settings to create the desired chat experience.

## How to Setup Project

Fill in the required fields and configure optional settings as needed.

1. Project ID\
   Enter a unique identifier for the project. This ID is used internally for system management and tracking.
2. Project Name\
   Specify the display name of the project. This name will appear in the user interface.
3. Description\
   Provide a brief summary of the project purpose or functionality.
4. Knowledge Base\
   Select one or more knowledge base to use for retrieving domain-specific information. You can add or manage knowledge bases from the Knowledge Base tab in the left sidebar.

#### **Advanced Settings**

<figure><img src="/files/lpt2MAPTV7cj1wUkH2yX" alt=""><figcaption></figcaption></figure>

1. Project Logo

   This section allows you to customize how the project displays its logo.

   1. Use Organization Logo\
      When enabled, the project will automatically use the same logo configured in user Organization Management settings.
   2. Show Application Logo\
      Enable this option to display the project logo on the New Chat page.
2. Agents
   1. Enable Agent Mode\
      When enabled, the project will operate in Agent Mode, meaning users will interact with an agent by default when opening the project. In Agent Mode, only one agent can be selected and assigned to the project. Once enabled, all user interactions within the project will be routed through the configured agent.
   2. Supported Agents\
      Select one or more agents to associate with this project. These agents will handle specific user inquiries and related use cases.
3. [Smart Hub Configuration](/glchat/organization-admin/how-to-guides/project/setup-project/smart-hub-configuration)\
   When **Smart Hub** is enabled, questions asked in the selected Project can be automatically routed to linked Projects. GLChat will then search the most relevant linked Project based on the user’s access permission.
4. RAG Pipeline\
   Select a Retrieval-Augmented Generation (RAG) **pipeline** to define how the system retrieves external information and integrates it into responses.
5. Preset ID\
   Choose a preset configuration that determines the application’s behavior. Available presets depend on the selected RAG pipeline.
6. Connector\
   Pick external connectors (apps or data sources) to integrate with your system. Only connectors registered under [Organization Management](/glchat/organization-admin/how-to-guides/integration-and-services/connector) are available.
7. [Question Library](/glchat/organization-admin/how-to-guides/project/setup-project/question-library)\
   Curate a **Question Library** of frequently asked questions (FAQs), either sourced automatically from the knowledge base or added manually.
8. [Project Discovery](/glchat/organization-admin/how-to-guides/project/setup-project/project-discovery)\
   **Project Discovery** controls whether a Project appears in the Project list for end users
9. [Prompt Builder Catalog Entries](/glchat/organization-admin/how-to-guides/project/setup-project/prompt-builder-catalog-entries)\
   Customize or extend default prompt templates to fit your use case. You can also enable syncing to automatically inherit updates from the master template.
10. [LMRP Catalogue Entries](/glchat/organization-admin/how-to-guides/project/setup-project/lmrp-catalog-entries)\
    Define system-level prompts for advanced or technical instructions. Similar to prompt templates, these can be synced to automatically apply master template updates.

After filling in the form, click Next to manage application access, or Cancel to discard the setup.

{% hint style="info" %}
Newly created project are **not automatically connected** to any knowledge base.
{% endhint %}


# Question Library

To access this features, go to the sidebar and select Project > New/Edit > Advanced Setting.

<figure><img src="/files/hqs13vCpdXohPYmTH43Y" alt=""><figcaption></figcaption></figure>

The Question Library allows admins to curate a list of frequently asked questions (FAQs) sourced from the project's knowledge base or entered manually.

These questions appear in the project UI as **recommended queries**, helping users quickly start a conversation, improve discoverability, and receive faster responses.

### Question Source Option

1. Manual Source\
   Admins can directly create and categorize questions that will appear in the project UI.
2. Knowledge Base Source\
   The system automatically generates recommended questions from the selected knowledge base. Admins only need to configure how these questions are displayed, how users interact with them, and how many are shown.

***

### Create Question Library

<figure><img src="/files/Sz6ngj7Yxedc6jtZ17U7" alt=""><figcaption></figcaption></figure>

Fill in the required information in the provided fields.

1. **Source**

   Defines how questions are populated:

   1. **Manual:** Admins manually create and manage the questions.
   2. **Knowledge Base:** Questions are automatically generated from the project’s existing knowledge base.
2. **Display Type**\
   Determines how the recommended questions appear in the UI:
   1. By Order: Questions appear in the order defined by the admin.
   2. Random: Questions are shuffled each time for variety.
3. **Click Behavior**\
   Defines what happens when a user clicks a recommended question:
   1. Send: The question is automatically sent to the LLM.
   2. Edit: The question is pre-filled in the input box, allowing the user to review or modify it before sending.
4. **Maximum Questions Shown on Project**
   1. Sets the upper limit for how many recommended questions are displayed to the user.
5. **Question Category (Manual Option Only)**
   1. A label used to organize questions into logical groups.
6. **Question Text (Manual Option Only)**
   1. Enter the question that will appear as part of the FAQ list and be shown to users as a suggested query.

\
\ <br>


# Prompt Builder Catalog Entries

To access this features, go to the sidebar and select Project > New/Edit > Advanced Setting.

<figure><img src="/files/GaEPwm6EJ7sG6gGYUuLw" alt=""><figcaption></figcaption></figure>

The **Prompt Builder Catalogue** is a collection of customizable prompt templates you can use with different AI models. This feature allows you to modify how your project responds by adjusting **system instructions** or **roles** for each prompt, ensuring greater control over tone, behavior, and model interaction.

***

#### **Catalogue Entry Details**

<figure><img src="/files/gEFwpqf9oeK3nTjRQgHJ" alt=""><figcaption></figcaption></figure>

Each prompt builder catalogue entry includes the following details:

1. Name\
   The identifier or title of the prompt template.
2. Grouping Type\
   Defines how the prompt is applied across models:
   1. No Grouping – The prompt is used by all available models.
   2. Scope – The prompt applies only to a specific model provider, a model name, or both.
   3. Template Group – The prompt applies only to a specific group of templates
3. Scope (Visible only if “Scope” is selected as the grouping type)
   1. Enter the model name or provider name to limit where the prompt will apply.
4. Template Group (Visible only if “Template Group” is selected as the grouping type)
   1. Select a template group from the predefined list of available templates groups.
5. System Instruction\
   Provide instructions that guide the overall behavior or tone of the project.\
   These are fixed rules or directives that shape how the system responds.
6. User Instruction\
   Define the instructions or inputs that users can provide during conversations to guide responses for specific interactions.

***

#### **View Prompt Builder Details**

To view details of a specific prompt builder catalogue entry, click the **View (eye icon)** in the **Action** column.

<figure><img src="/files/x3RYZ8FBkGPPx3Oe8IYK" alt=""><figcaption></figcaption></figure>

***

#### **Edit Prompt Builder Details**

To edit details of a specific prompt builder catalogue entry, click the **Edit (penci icon)** in the **Action** column.

<figure><img src="/files/GmbGcTU9HmlUS7y6kbNj" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
If you want to edit the entry, ensure that the **Sync Template** toggle is **off**.\
When the sync template is **on**, the entry will be locked in **view-only mode** to maintain consistency with the master template.
{% endhint %}

***

#### **Sync Template**

The **Sync Template** feature keeps your prompt catalogue synchronized with the setting on [Prompt Builder](broken://pages/ZE2Xcqn0rXOEziqwgiRl) menu, ensuring your system always uses the latest standardized prompts and configurations defined at the global level.

You can toggle synchronization on or off depending on whether you want to inherit updates or customize locally.

| **Mode**                       | **Behavior**                                                                                                                                              |
| ------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Sync Template (Toggle On)**  | Automatically inherits all updates from the master template. Manual editing is disabled to preserve consistency with the master version.                  |
| **Sync Template (Toggle Off)** | Converts into an **Override Template**, allowing full customization. However, changes will not update automatically when the master template is modified. |


# LMRP Catalog Entries

To access this features, go to the sidebar and select Project > New/Edit > Advanced Setting.

<figure><img src="/files/v1wez2EFqrClgkCD9u9c" alt=""><figcaption></figcaption></figure>

The **LMRP (Language Model Request Processor) Catalogue** defines system-level prompts used to control and fine-tune how the project processes and structures responses.\
These entries provide advanced or technical instructions that guide the AI’s internal operations — such as transforming queries, formatting outputs, or generating conversation titles.

***

#### **Catalogue Entry Details**

<figure><img src="/files/4y8oizCIWe4uBXJ9Yfwn" alt=""><figcaption></figcaption></figure>

Each LMRP catalogue entry includes the following details:

1. Name\
   The identifier or title of the prompt template.
2. Scope
   1. Enter the model name or provider name to limit where the prompt will apply.
3. Prompt Builder System Template\
   Provide instructions that guide the overall behavior or tone of the project.\
   These are fixed rules or directives that shape how the system responds.
4. Prompt Builder User Template\
   Define the instructions or inputs that users can provide during conversations to guide responses for specific interactions.
5. LMRP Model\
   References a model created under the LMRP Model section in the LLM Model menu.\
   This links the catalogue entry to the appropriate processing model.
6. Output Parser Type\
   Determines how the chatbot formats its responses:
   1. JSON – Use for structured outputs.
   2. None – Use when no specific output format is required.

#### **View LMRP Catalog Entries Details**

To view details of a specific LMR catalogue entry, click the **View (eye icon)** in the **Action** column.

<figure><img src="/files/ysUWSiZXll8ZXoOY1WXF" alt=""><figcaption></figcaption></figure>

***

#### **Edit LMRP Catalog Details**

To edit details of a specific **LMRP Catalog** catalogue entry, click the **Edit (penci icon)** in the **Action** column.

<figure><img src="/files/PTqfsp4g7thruilPxhEj" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
If you want to edit the entry, ensure that the **Sync Template** toggle is **off**.\
When the sync template is **on**, the entry will be locked in **view-only mode** to maintain consistency with the master template.
{% endhint %}

***

#### **Sync Template**

The **Sync Template** feature keeps your prompt catalogue synchronized with the setting on [LMRP Catalogue](broken://pages/2wS3uzGNh8hWH2ZToLuE) menu, ensuring your system always uses the latest standardized prompts and configurations defined at the global level.

You can toggle synchronization on or off depending on whether you want to inherit updates or customize locally.

| **Mode**                       | **Behavior**                                                                                                                                              |
| ------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Sync Template (Toggle On)**  | Automatically inherits all updates from the master template. Manual editing is disabled to preserve consistency with the master version.                  |
| **Sync Template (Toggle Off)** | Converts into an **Override Template**, allowing full customization. However, changes will not update automatically when the master template is modified. |


# Smart Hub Configuration

### Overview

Smart Hub can be configured from the **Admin Dashboard** at the Project level.

When Smart Hub is enabled, questions asked in the selected Project can be automatically routed to linked Projects. GLChat will then search the most relevant linked Project based on the user’s access permission.

This allows the Project to act as a central entry point for users.

***

### Smart Hub Setting

In the Project configuration page, admins can enable or disable **Smart Hub** using the toggle.

When enabled, the Project will behave as a Smart Hub Project.

<figure><img src="/files/6NRSx8FPlbYGz2izPaOF" alt=""><figcaption><p><strong>Smart Hub configuration in Project settings</strong></p></figcaption></figure>

***

### Linked Projects

**Linked Projects** are the Projects connected to the Smart Hub Project.

When users ask a question from Smart Hub, GLChat will check the linked Projects and search only the Projects that the user is allowed to access.

For example, if Smart Hub is linked to:

<table><thead><tr><th width="334.00006103515625">Linked Project</th><th>Knowledge Base</th></tr></thead><tbody><tr><td>Project 1</td><td>HR policies</td></tr><tr><td>Project 2</td><td>Finance policies</td></tr><tr><td>Project 3</td><td>IT support</td></tr></tbody></table>

A user can ask from Smart Hub, and GLChat will automatically search the relevant linked Project based on the user’s access.

***

### How to Configure Smart Hub

1. Open the **Admin Dashboard**.
2. Go to the Project that will be used as the Smart Hub Project.
3. Enable the **Smart Hub** toggle.
4. Select the Projects that should be linked to Smart Hub.
5. Save the configuration.

After configuration, users can ask questions from the Smart Hub Project without manually switching between linked Projects.

***

### Important Notes

Smart Hub does not give users access to every linked Project.

GLChat still checks each user’s access permission before searching a linked Project. If a user does not have access to a linked Project, that Project will not be searched for that user.


# Project Discovery

### Overview

**Project Discovery** controls whether a Project appears in the Project list for end users.

This is useful when a Project is mainly used as a linked Project under Smart Hub and does not need to be directly selected by users.

***

### Hide from Project List

Admins can use the **Hide from Project List** setting to hide a Project from the end-user Project list.

When enabled, the Project will not appear in the Project list for end users.

<figure><img src="/files/KtNJAPgZ587rGQxo2G9m" alt=""><figcaption><p><strong>Project Discovery setting to hide a Project from the end-user Project list</strong></p></figcaption></figure>

***

### How Project Discovery Works with [Smart Hub](/glchat/organization-admin/how-to-guides/project/setup-project/smart-hub-configuration)

Project Discovery only controls whether a Project is visible in the user interface.

It does not remove the user’s access permission, and it does not prevent Smart Hub from using the Project as a linked Project.

For example:

| Setting                                   | Result                                               |
| ----------------------------------------- | ---------------------------------------------------- |
| Project is visible                        | User can see and open the Project directly           |
| Project is hidden                         | User does not see the Project in the Project list    |
| Project is hidden but linked to Smart Hub | Smart Hub can still search it if the user has access |

***

### Example Use Case

An organization has several Projects:

| Project           | Purpose                            |
| ----------------- | ---------------------------------- |
| Smart Hub Project | Main entry point for users         |
| HR Project        | Contains HR knowledge base         |
| Finance Project   | Contains Finance knowledge base    |
| IT Project        | Contains IT support knowledge base |

Admins can make the **Smart Hub Project** visible to users, while hiding the HR, Finance, and IT Projects from the Project list.

Users only need to open Smart Hub and ask questions there. Behind the scenes, GLChat will still search the linked Projects based on the user’s access.

***

### Recommended Setup

For a simpler user experience:

<table><thead><tr><th width="335.77783203125">Project Type</th><th>Recommended Visibility</th></tr></thead><tbody><tr><td>Smart Hub Project</td><td>Visible</td></tr><tr><td>Linked sub-projects</td><td>Hidden, if users do not need to open them directly</td></tr></tbody></table>

This helps reduce Project list clutter while still allowing Smart Hub to retrieve answers from the right knowledge bases.

***

### Important Notes

Hiding a Project from the Project list is not the same as removing access.

A hidden Project can still be used by Smart Hub if:

1. The Project is linked to Smart Hub.
2. The user has permission to access that Project.
3. The Project contains relevant knowledge for the user’s question.

***


# Manage Access

<figure><img src="/files/ntmaXwr5TKSl7zqQlqww" alt=""><figcaption></figcaption></figure>

Use the **Manage Access** section to control who can access a project and what level of permission they have. There are two types of access you can manage:

* **Project Access:** Grant end users access to an existing project. This determines which users can interact with or use the project.
* **Dashboard Access:** Manage admin-level access to the project dashboard. Admins can be given permission to view or edit the project, depending on their assigned role.

{% tabs %}
{% tab title="Project Access" %}

<figure><img src="/files/ntmaXwr5TKSl7zqQlqww" alt=""><figcaption></figcaption></figure>

Use the **Project Access** tab to grant users access to an existing application. This tab becomes available only after the application setup is complete.

#### Public Access

<figure><img src="/files/1T7M3MqZIEReRVEXwGUL" alt=""><figcaption></figcaption></figure>

Use **Public Access** to allow guests who are not logged in to GLChat to access the project. When enabled, choose one of the following access types:

1. **Visitor**
   * Allows access for:

     1. Guest visitors who are not logged in
     2. Logged-in users added in **Permitted Access**

     This option is recommended when the project should be publicly accessible to visitors, while logged-in user access remains limited to selected users, teams, or domains.
2. **Visitor and Logged-in Users**
   * Allows access for:

     * Guest visitors who are not logged in
     * Any logged-in GLChat user

     When this option is selected, **Permitted Access** is hidden because all logged-in users automatically have access.

#### Permitted Access

<figure><img src="/files/X3hXmDiW6ayRH5mrU44G" alt=""><figcaption></figcaption></figure>

The **Permitted Access** section is available only when **Visitor** is selected. Use this section to grant access to specific logged-in users while restricting other authenticated users.

{% hint style="info" %}
Before assigning access, make sure the user, team, or domain has already been added in [User Management](/glchat/organization-admin/how-to-guides/user-management).
{% endhint %}

To grant access:

1. Click **Assign User**.
2. Choose one of the following access options:
   * **Account**: Grant access to individual users.
   * **Team**: Grant access to all members of a selected team.
   * **Organization Domain**: Grant access to users with a matching email domain, such as `example.com`.
3. Click **Save**.

{% hint style="info" %}
The project creator automatically has full access to the project.
{% endhint %}
{% endtab %}

{% tab title="Dashboard Access" %}

<figure><img src="/files/L59gRF4YcagDVeXAJdVw" alt=""><figcaption></figcaption></figure>

Use **Dashboard Access** to manage admin permissions for the project dashboard.

{% hint style="info" %}
By default, the user who creates the project becomes the **Owner**. The owner can s**hare, edit, and delete the project.**
{% endhint %}

You can add other admin users and assign one of the following roles:

* Viewer: Viewers can access and view the project dashboard but cannot make changes.
* Editor: Editors can edit the project and manage sharing access. They can also add new users to the project.

#### Project List View by Role

The project list view may differ depending on the user’s assigned dashboard role.

**Owner**

Owners have full access to manage the project. They can view, edit, share, and delete the project.

<figure><img src="/files/36JgAzMvYRnacNA3diD7" alt=""><figcaption></figcaption></figure>

**Editor View**

Users with the **Editor** role can edit the project and managing sharing access. The editor view will look like this:

<figure><img src="/files/tgYkfJx57nWaIiG4J81v" alt=""><figcaption></figcaption></figure>

**Viewer View**

Users with the **Viewer** role can view the project but cannot make changes or access editing actions. The viewer view will look like this:

<figure><img src="/files/X9cfSCOXN9CCr6OafEiJ" alt=""><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}


# Manage FAQ

<figure><img src="/files/ah3ZgPxpgdSBHOigtSw6" alt=""><figcaption></figcaption></figure>

This feature enables admins to create and manage a list of **frequently asked questions (FAQs)** derived from the GLChat project's knowledge base. The FAQs appear in the [**Question Library**](/glchat/application-ui/how-to-guide/chat-input-and-responses/question-library) of the associated GLChat project.

{% hint style="info" %}
A project can contain multiple FAQs, and these FAQs may originate from one or more Knowledge Bases associated with the project.
{% endhint %}

***

### Select Knowledge Base

To manage To manage FAQs, first select the desired **Knowledge Base** from the dropdown menu. The FAQ list and settings displayed will automatically update based on the selected Knowledge Base.

Each Knowledge Base has its own FAQ quota, which can be configured independently through "[Settings](#setings)"

<figure><img src="/files/aucR0C1jR9FoY8WipTCA" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
Only Knowledge Bases that are currently connected to the project (refer to field "Knowledge Base" in [Setup Project](/glchat/organization-admin/how-to-guides/project/setup-project))will appear in the dropdown list.
{% endhint %}

***

### Auto Generate FAQ

This feature allows the system to automatically generate FAQs from the selected Knowledge Base on a scheduled basis.

<figure><img src="/files/a4GKTVybNhLVoJPeUe9x" alt=""><figcaption></figcaption></figure>

The generation process runs nightly between **11:00 PM and 4:00 AM (GMT+7)**. During each run, the system generates FAQs based on the configured quota for the selected Knowledge Base.

{% hint style="info" %}
If the generation process is not completed within the scheduled window, it will automatically resume during the next scheduled run until the quota is fulfilled.
{% endhint %}

***

### Add New FAQ Manually

<figure><img src="/files/lPmeIwF2fZ8LnifXczUC" alt=""><figcaption></figcaption></figure>

To create a new FAQ, fill in the required fields and click **Next** to proceed or **Cancel** to discard changes.

1. **Question**\
   Enter the question that will be displayed to users.
2. **Knowledge Base**\
   Select one or more documents from the available knowledge base to use as reference materials. If no document is selected, the system will automatically use all documents in the knowledge base.
3. **Answer**\
   You can enter the answer manually, or click **Generate Answer** to have the LLM create it automatically.
4. **Category**\
   Assign a category to help organize and group related questions more easily.

***

### Activate/Deactivate FAQs

By default, newly generated FAQs go through a moderation process before being displayed in the [Question Library](/glchat/application-ui/how-to-guide/chat-input-and-responses/question-library). Administrators can review and choose whether to activate or deactivate each FAQ.

<figure><img src="/files/RNqB3M2PTm6rjcb7y0Mo" alt=""><figcaption></figcaption></figure>

* **Activate FAQ** – The question becomes visible in the **Question Library** and can be accessed by project users.
* **Deactivate FAQ** – The question is hidden from the **Question Library** but remains available in the admin panel and can be reactivated at any time.

Below is an example of how auto-generated FAQs are displayed on the project home page:

<figure><img src="/files/FU5nJZosfCukmrC4rZJ3" alt=""><figcaption></figcaption></figure>

***

### Export Logs

To export FAQ activity logs, click the arrow next to **New FAQ** and select **Export Logs**.

<figure><img src="/files/Jp4Cj2YbD9ahDGlSdUdc" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
The exported file contains all available FAQ questions based on the selected filter criteria.

* If no date filter is applied, the system exports all available questions.
* If a date range is specified, only questions created within the selected period will be included in the export.
  {% endhint %}

***

### Setings

To configure FAQ settings, click the arrow next to **New FAQ** and select **Settings**.

<figure><img src="/files/u9M1hRVQYxDH2p4Fui1S" alt=""><figcaption></figcaption></figure>

The settings modal will open:

<figure><img src="/files/B6028CaxQVOrs7lFVIu7" alt=""><figcaption></figcaption></figure>

Admin can configure the following options:

* **Cache Time-to-Live (TTL)**

  Defines how long generated FAQ responses remain cached for each Knowledge Base.
* Maximum Number of Questions

  Determines the maximum number of FAQs that can be displayed in the Question Library for each Knowledge Base. Once the configured limit is reached, the New FAQ button will be disabled until existing FAQs are removed.


# Channel

<figure><img src="/files/XfLueOARJdoDBcM55uZs" alt=""><figcaption></figcaption></figure>

Define how and where users interact with GLChat. Select your preferred **channel** and **interface type** from the available options:

1. Web Widget Embed the chatbot into an existing website as a **Floating Widget** (corner button) or a **Half-Page Widget** (expanded view). For detailed implementation instructions, please refer to the [**GLChat Widget Setup**](/glchat/developer-documentation/glchat-widget-setup) section.
2. Standalone Page Host the chatbot as a dedicated, full-page web application.
3. WhatsApp Connect GLChat to a WhatsApp Business account via API integration, allowing users to chat directly through WhatsApp. Before enabling **WhatsApp** as a deployment option, you must first register your **phone number** through the[ **Communication Provider**](/glchat/organization-admin/how-to-guides/integration-and-services/messaging-channel) **menu**.

After selecting the deployment option, click **Save** to complete setup or **Previous** to return to the previous step.




---

[Next Page](/glchat/llms-full.txt/1)