General Assistant

The General Assistant is designed to streamline finding information across your company’s internal and external tools. By creating bots equipped with customized knowledge bases and deploying them to your tools, users can effortlessly pose queries and promptly receive pertinent responses from company or external knowledge, making it easier for everyone to get the knowledge they need when and where they need it, whether in the Unleash chat or within your knowledge touchpoints.

You can create many Assistant bots and select the Assistant bot of your choice when connecting it to your tools, ensuring tailored and efficient response management.

Step 1 | Creating a General Assistant

To create a general assistant:

• Click on Control Center.

• Click on Assistants.

• Click +New Assistant.

• Click General Assistant.

Note: If you're creating a general assistant to support the chat page, you'll need to share it with the relevant users you want it to assist. Learn more about sharing assistants.

Step 2 | Configuring the Assistants Source

Define the knowledge source for your assistant to suit your needs:

Knowledge Source

• Connected Sources: Leverage your internal knowledge base for precise, company-sourced answers. This option uses the GPT-4o model by default.

  • When selecting company knowledge, a new section will open, allowing you to configure the assistant’s identity and knowledge sources.

• Public Sources: Tap into external, public AI models for broader, general knowledge. Choose from a range of powerful options outlined in the table below.

  • Note: You can only select one LLM, and the model cannot be changed once the assistant has been saved.

  • When selecting GPT-5 or Gemini 3, you will need to choose a reasoning level from the dropdown that appears.

How to choose your LLM

Connected Sources

Identity

By default, the assistant’s identity is set to the user who created it. The assistant operates under that identity’s data access permissions. If the identity is changed, the assistant’s connected data sources and permissions will update to match those of the new identity. The assistant can access only the data sources that the chosen identity is authorized to use. As a result, switching the identity may change which data sources are available.

Users with full access to the assistant, as well as admins, can edit the assistant’s connected data sources. However, the data sources that can be added or configured are determined entirely by the assistant’s selected identity, not the person making the edits.

When someone interacts with the assistant, the responses it provides are based on the data-source permissions of the person asking the question.

Data Sources

To configure the assistant’s sources:

• Click on + Add App.

• Select the relevant app you want to connect.

  • If you have multiple links connected for an app, a popup of the links you have access to will appear. Click on the relevant account.

• To add resources from Unleash, select Unleash.

  • Under Wikis, choose the wiki/s you would like to include.

  • You can add Unleash again as an app and select additional wikis.

• Narrow the scope of the resources the bot will use by clicking on +More Filters (optional).

• As you apply filters, the system dynamically updates the available options, ensuring that new filters remain within the scope of your current selection. Three constants are: Specific Resource, Last Edited and Type.

  Specific resource:

  • Search for a resource and click the checkbox next to the relevant resource to check specific resources as data sources.

  • You can select as many as you want.

  • Once done, click Select.

  Last edited: Establish specific timeframes for the bot to draw answers from when responding to a question.

  • Choose a time filter or scroll to the bottom and select Custom.

  • Select a time from Before, After, or Between.

  • Select the dates.

  • To save, click out.

  Type:

  • Select the type of resource available for that app.

• To add another app, click +Add App (optional).

  • You can activate knowledge prioritization when you have more than two app sources configured. The assistant will seek answers within the prioritized resources you set. If no answer is found, it will expand its search to other connected resources.

    To activate knowledge prioritization:

    • Click Knowledge Prioritizationon the sources to prioritize (you can prioritize more than one).

• Click Save.

Link Indications

• If any card contains a file, the Assistant will scan and utilize those files to provide more accurate answers

• Private wikis and links that are set as a data source will be visible as answers delivered by the bot in channels. Private wikis and links are identifed by this icon:

• Links that are still syncing to Unleash are displayed with this syncing icon:

• If the selected link fails to sync, an indication will appear below the app, prompting you to try reconnecting the link on the Unleash app connection page.

• If the link was removed, an indication will appear below the app.

• When a link is selected within 24 hours of its initial connection, a message informing you that full AI functionality will be available within 24 hours of the connection time will be displayed.

Public Sources

Files

This feature lets you upload files directly into your LLM Assistant’s private knowledge base, allowing it to generate more accurate, relevant, and organization-specific answers, going beyond its general training data.

What happens when you upload files?

• Once uploaded, your documents become part of the Assistant’s private knowledge base.

• The Assistant can reference, summarize, and answer questions directly based on your uploaded content.

• You can preview or delete uploaded documents at any time.

Note: If the file upload toggle is turned off, the assistant will not use any files from this section when searching or generating answers.

How does the Assistant use uploaded files?

• When you ask a question, the Assistant first searches your uploaded knowledge base.

• If it finds a match, the AI will answer only using that content.

• If no match is found, it will fall back to:

  • Web search (if enabled)

  • Its general training data

File upload requirements

• Supported Models: Available for Assistants using GPTs (support for additional models coming soon).

• Upload Limit: Up to 20 files per Assistant.

• Maximum File Size: 50 MB per file and 20MB for images.

• Permissions: Only users with Full Access to the Assistant can upload documents.

Step 3 | Configuring the Assistant Tools (Public Source Only)

File Upload

The file upload feature allows you to send files as messages to LLM-configured general assistants via the web app, desktop app, or browser extension. This capability helps the assistant generate more accurate and relevant answers.

To enable this feature, toggle on the File Upload setting.

Note: Claude 3.7 Sonnet and Claude 3.5 Haiku do not support file uploads, so the tool will be disabled for them.

Note: If the this toggle is turned off, the assistant will not use any files from the Files section when searching or generating answers.

Web Search

When asking a question that requires up-to-date, time-sensitive information not included in the LLM's built-in knowledge (e.g., today's weather or the current exchange rate), the LLM can provide web-based results.

To enable this functionality, ensure the Web Search setting is toggled on.

Step 4 | Configuring the Assistant Instructions

The instructions are set to the default settings, allowing you to configure the response length and adjust the tone. The advanced settings button enables you to customize and control the assistant’s behavior, tone, and response style by giving free text instructions.

Default Instructions

Response Length

Configure the response length that the assistant will use to deliver answers. You can set the response length to short, standard, or long, with the default being standard.

To configure the response length:

• Click on the dropdown for Response Length.

• Select:

  • Short: Typically consists of 1-2 sentences, providing quick, straightforward answers.

  • Standard: Slightly longer, usually a paragraph or a few sentences, offering more detailed explanations.

  • Long: In-depth and comprehensive, often spanning multiple paragraphs or containing detailed explanations.

Adjust Tone

Configure the tone that the assistant will adopt when delivering answers. You can set the tone to neutral, assertive, and confident, with the default being neutral.

To adjust the assistant’s tone:

• Click on the dropdown for Adjust Tone.

• Select:

  • Neutral: This is the default tone. It mirrors the language style documented in your resources and provides a balanced and professional response.

  • Assertive: This tone is direct and clear, communicating the necessary information with authority. It is used when conveying a strong stance.

  • Confident: This tone is positive and assured, instilling confidence in the customer. It is used when reassuring the customer or demonstrating expertise.

Note: Switching between standard and advance resets the previous values without saving them.

Advanced Instructions

The Advanced Settings button enables you to personalize the assistant’s behavior, tone, and response style by entering custom free-text instructions.
You can provide detailed instructions up to 16,000 characters. For LLM assistants, the limit is 32,000 characters. For best practices, please see this article.

You can provide instructions to tailor:

• Tone (e.g., formal, casual, assertive, confident).

• Response Length (e.g., concise, moderate, detailed).

• Role and Behavior Guidelines (e.g., acting as a professional customer support representative or a knowledgeable expert in a specific field).

Connected Sources Example: As a product agent in Acme's product chat, your role is to provide clear and concise answers about our product features, updates, and usage. Respond promptly, tailoring your explanation to the user's level of familiarity with the product. Use a friendly and professional tone, and include links to relevant documentation or resources when needed.

LLM Example: As an external LLM chat agent representing Acme, your role is to provide accurate, clear, and helpful responses to users seeking information. Focus on understanding user queries and delivering concise, informative answers. Maintain a neutral and professional tone while ensuring your responses are easily understood by users of varying expertise levels. If applicable, suggest further resources or actions based on the user’s needs.

Step 5 | Configuring the Assistant Promt Templates

Prompt Templates

Add prompt templates to your assistant to help workspace members understand what questions they can ask the assistant in the chat. These prompts serve as clear, practical examples that guide users and encourage engagement. For example:

• What are our company policies on remote work?

• How do I request vacation leave?

• What does error [error name] mean?

• What was the latest update on [customer name]?

When a member clicks a prompt template card from the chat’s empty state, the prompt text is inserted into the input field for easy editing before sending. Learn more about using prompt templates in the chat.

To create a prompt template:

• Click inside the prompt field and type your message.

• Use brackets to mark editable variables (e.g., How many [subject] did we close in [time]?).

• Prompts are saved automatically when you click outside the field.

• To edit a prompt, simply click on it.

• Click + Add to add another prompt.

Note: Each assistant can have up to four prompt templates only.

Step 6 | Configuring the Assistant Display Preferences

There are two important toggles to optimize how users interact with the General Assistant. By default these toggles are off.

Pin to Chat Sidebar: Displays the assistant in the chat sidebar for users with interact permissions, allowing users to start a new chat with the assistant easily. Only users with Full Access can unpin this assistant.

Display in Quick Search: Integrates the assistant into the quick search experience. When typing a query into the homepage search bar or launch bar, a dropdown will appear, enabling users to select the chat assistant to answer their questions or receive answers from the Unleash Workspace Assistant. Learn more here.

Step 7 | Deployment in Slack

This section is only visible to admins. This allows admins to create an AI agent based on the assistant to chat with in Slack. The Deploy in Slack button remains disabled until the assistant is saved.

To deploy the assistant in Slack:

• Click Save.

• Click Deploy in Slack.

• Follow the steps from this guide to take the next steps in creating the AI Agent.