⚒️How to build a Knowledgebot

Building a Knowledge Bot is your first step towards creating an AI-powered assistant tailored to your specific needs. This process transforms your data and documents into an interactive, intelligent system that can answer questions, perform tasks, and support your team's daily operations.

Please note that not all steps are required for a quick bot setup. Specifically, steps 3, 4, 5, 6, and 9 are optional and delve deeper into the details. These steps can be configured at a later time if needed. For a quick bot setup, simply follow steps 0, 1, 2, 7, 8, and 10.

Step 0 - Create a bot in the 'Workshop'

Go to your Workshop dashboard. Choose whether you want to build a Nexus or a Retriever. Then Click "Create +".

Here you will configure the bots you want to share with others in your organization.

Step 1 - Name the bot & chose an avatar

Begin by giving your bot a unique and memorable name. This is how users will identify your bot. You can also upload an Avatar so your bot may also be identified visually.

Step 2 - Provide Instructions

Define your bot's core mission, writing style, and any context it should be aware of. Think of it like a brief job description that the bot will always consider in its responses. Use the default template or write your own instructions to customize the bot to fit your use case. For most business use cases, a simple and more generic prompt can perform better than long and complex descriptions.

This is the Initial instruction prompt we recommend:

Your task is to provide precise and relevant information. Your communication should be professional yet easy to understand. Your responses must be thorough and detailed, using professional formatting. Clarify Ambiguous or General Inquiries: Ask specific questions to gather additional context. If a query can have multiple interpretations or answers, ask the user for clarification. Inform Users About the Scope of Answers: Let users know that answers may not always be exhaustive, indicating that further research and verification might be necessary. Encourage users to consult the original documents in the reference if they wish to look up more information themselves, or they can ask follow-up questions for further assistance. Methodical Approach: Guide users through the answer to their question methodically. Clearly state the steps to be taken and, where possible, refer to specific sections, figures, and tables in the guidelines and forms. Your ultimate goal is to enhance the expertise and efficiency of users, enabling them to perform their tasks more effectively through quicker access to relevant information.

If you want the Knowledge Bot to behave in a more specific way and limit the scope of its tasks, we recommend to include more context in the instruction prompt, expanding on the default template:

Role/Context: Provide the bot with your unique business context and needs.

• e.g. "You are assisting the tech staff with internal IT support."

Task: Describe the bot's main functionality in a concise way.

• e.g. "Provide factual responses and step-by-step guides using the provided knowledge base."

For most use cases where the bot's responses are based on a typical knowledge base, the default prompt does not need to be changed.

Tonality: Tailor the writing style to your use case and user group.

• e.g. "You write in a professional, yet easy to understand tone. Your answers are very precise and brief. You use professional formatting, like headlines, sub-headlines, bold, italic, underline, or numbering."

Note: When the bot's main purpose it to create content in a certain tone, it is advisable to upload branding guidelines and past content pieces into a knowledge base or into the context as references to teach the bot the required writing style.

• e.g. "You write in the same style and tonality as the documents provided by the user."

Guidelines: It is recommended to set guidelines that minimizes the likelihood of hallucinations.

• e.g. "Do not make up an answer when you cannot offer a factual answer based on your knowledge base. Tell the user what information you are missing to respond."

Step 3 - Create a "Welcome Message"

The welcome message is your bot's first interaction with users and sets the tone for all future engagements. Navigate to Bot Settings to begin crafting your greeting. Your message should introduce your bot's purpose while outlining its key capabilities. Include clear instructions for getting started and set realistic expectations about what the bot can and cannot do. A well-crafted welcome message guides users naturally into their first interaction and helps them make the most of your bot's capabilities.

Pro Tip: Test your welcome message with new users to ensure it effectively guides them into their first interaction with your bot.

Step 4 - Add a Description to your Bot

A clear and comprehensive bot description helps users understand your bot's capabilities and intended use. Navigate to Bot Settings to create or edit your bot's description. Your description should explain the bot's primary purpose, its area of expertise, and the types of tasks it can handle. This information appears in your bot's profile and helps users decide if this bot matches their needs. Keep the description concise but informative, focusing on practical use cases and specific features that set your bot apart.

Note: A well-written description can significantly improve user engagement and ensure appropriate use of your bot.

Step 5 - Conversation Starters & Follow-Up Questions

Conversation starters and follow-up questions help guide users through meaningful interactions with your bot. Navigate to Bot Settings > Conversation Design to configure these elements.

Conversation Starters

Conversation starters are pre-defined questions or prompts that users can click to begin their interaction. Create starters that showcase your bot's key capabilities and common use cases. These appear as clickable suggestions when users first engage with your bot, eliminating the guesswork of how to begin.

Follow-Up Questions

Follow-up questions maintain conversation flow and deepen user engagement. These dynamic suggestions appear after your bot's responses, encouraging users to explore related topics or get more detailed information. Design follow-ups that naturally extend from previous answers and guide users toward valuable insights.

Pro Tip: Regularly review and update your conversation starters and follow-ups based on actual user interactions and feedback.

Step 6 - Capabilities & Skills

Audio Features

Text-to-Speech Output

Configure your bot's voice output by enabling OpenAI's text-to-speech capability. Select from a range of pre-defined voices and use the sample playback feature to find the perfect voice for your bot's personality and purpose.

Speech-to-Text Input

Enable voice command functionality to allow natural spoken interactions with your bot. Users can speak their queries directly, creating a more intuitive and accessible experience.

Image Processing

Enhance your bot's visual capabilities with Smart Image Recognition and Image Retrieval features. These are crucial when working with image-based content in your databases and documents, allowing your bot to understand and reference visual information effectively.

Citation Settings

The Inline Citations toggle determines how your bot references its sources. When enabled, it seamlessly integrates source citations directly within responses, providing transparency and credibility to your bot's answers.

Search Methods

1. Index Search Traditional full-text search provides precise matching for straightforward queries. This method excels at finding exact matches within your content.

2. AI Search Semantic search capabilities enable your bot to understand context and meaning beyond exact word matches. This method is particularly effective for complex, nuanced queries

3. Hybrid Search A sophisticated combination of full-text and semantic search, offering the best of both worlds. This method balances precise matching with contextual understanding, making it ideal for diverse use cases.

Note: Choose your search method based on your specific needs - Index Search for precision, AI Search for context understanding, or Hybrid Search for versatile applications.

Step 7 - Choose a LLM for your Bot + Model Modifier

LLM Selection

Language Learning Models (LLMs) power your Knowledge Bot by processing queries, understanding context, and generating responses. Choosing the right LLM can enhance accuracy, performance, and cost efficiency based on your needs.

Select from leading LLMs provided by Azure, Vertex, OpenAI, and Anthropic, each offering different capabilities, including:

• Context window size

• Response quality

• Performance

• Cost efficiency

• Hosting location

• Specialized capabilities

Select your preferred model with a simple click based on your specific requirements.

Model Modifiers

Fine-tune your bot's behavior with these essential parameters:

Creative Freedom

Controls the balance between creativity and consistency:

• Higher values: More creative, varied, but less predictable responses

• Lower values: More factual and consistent outputs

Vocabulary Range

Adjusts the breadth of language used:

• Higher values: More diverse vocabulary, potentially less relevant

• Lower values: More focused vocabulary, higher relevance

Topic Variety

Manages the introduction of new ideas:

• Higher values: More varied topics and fresh perspectives

• Lower values: More focused discussion, higher coherence

Word Variety

Controls vocabulary repetition:

• Higher values: Broader word choice, fewer repetitions

• Lower values: More familiar vocabulary, consistent terminology

Search Range

Defines the range of information considered:

• Higher values: Broader document search, more comprehensive but potentially less precise

• Lower values: Focused search, higher relevance but narrower context

Note: Adjust these parameters based on your specific use case and desired bot behavior.

Step 8 - Connect a Knowledge Base

To train the bot on your business context and use case, you'll need to equip it with relevant documents. Connect your bot to existing databases or create new knowledge bases to enhance its capabilities. Access this feature through the Database tab in your bot settings.

Accessing Databases

View a complete list of available databases on your instance. Simply click on any database to connect it to your bot.

Creating a New Knowledge Base

Basic Setup

1. Click "Create Database" located in the top right corner, this button starts the setup process.

2. Enter the required information:

   • Database name – Choose a clear, descriptive name for easy identification.

   • Description – Provide a summary of the database's contents and purpose. This may help improve the AI's accuracy especially with more complex Knowledge Bases.

   • Embedding Model selection – Choose an embedding model based on your needs. Consider factors like search accuracy, language support, and performance when selecting the best model. (For more details, visit the Advanced Features page.)

Advanced Settings

Access Advanced Settings via the button in the bottom left - don't miss this crucial step!

Configure additional parameters:

• Access Settings (Private/Public) – Control who can view and contribute to the database.

• Character Chunk Size – Defines how much text the AI processes in a single chunk. Smaller chunks provide more precise responses, while larger chunks capture broader context. You may opt to change the default settings.

• Chunk Overlap – Determines how much text overlaps between chunks to maintain continuity in AI analysis. Higher overlap improves context retention, lower overlap can lower processing time.

• Smart Table Processing – Converts tables in PDFs into structured text, making them AI-readable. (Enabling this may increase processing time and cost.)

• Smart Image Processing – Extracts text from images in PDFs, allowing AI to interpret and use visual data. (This also affects processing costs.)

Note: Enable Smart Table and Image Processing if your database contains tables or images that need analysis.

Step 9 - Setting Up Agents & Workflows

Agents

Agents are reusable, customizable prompt shortcuts that help streamline tasks, saving time and improving productivity. They allow you to automate specific queries, ensuring consistency and efficiency in responses.

Managing Your Agents

The Agents tab gives you access to your Agent Library, where you can browse and manage pre-configured agents for different tasks.

• On the left side, explore various agent categories.

• On the right side, view available agents within each category, along with a description of their purpose and functionality.

In the Source tab, you can:

• Edit prompts to customize an agent’s behavior.

• Delete agents that are no longer needed.

• Mark frequently used agents as favorites for quick access.

To add an agent to your bot, click the plus (+) icon next to the desired agent, and it will automatically move to the Selected tab, integrating with your bot.

Creating Custom Agents

Create new agents using the "Add Agent Task" buttons in the top-right corner, allowing you to build custom solutions for your specific needs. To create a new agent:

1. Title – Choose a clear, descriptive name for the agent. This will act as the button name for the prompt shortcut.

2. Instructions – Provide a detailed explanation of what you want the agent to do. Think of this as constructing a well-defined prompt for the AI.

3. Description – Add a short summary explaining the agent's purpose. This helps your team understand its function at a glance. (Tip: You can use Blockbrain’s built-in templates to simplify the process.)

For more ideas on how to use agents, visit the Advanced Features page.

Workflows (Automations)

Workflows allow you to automate multi-step prompts, streamlining complex interactions for more structured and efficient conversations.

Creating New Workflows

Start by clicking the New Workflow button on the left side. Access workflow settings through the three-dot menu in the top-right corner of the New Workflow box, where you can edit the title and description of your automation sequence.

Workflow Configuration

The right window displays your workflow steps as individual tabs. Add new steps using the plus icon and arrange them in your preferred order using drag and drop functionality. For each step, configure the name, instructions (prompt), and select the specific LLM model to be used.

Execution Control

Choose between two execution modes for your workflow:

1. Autopilot Mode

• Fully automated execution

• Runs through all steps without interruption

• Best for standardized processes

• Maximum efficiency

2. Human-in-the-Loop Mode

• Requires manual approval between steps

• Offers greater control over the process

• Can be configured globally or per step

• Ideal for sensitive tasks requiring oversight

For use case ideas and best practices, check out the Advanced Features page.

---

Intent Agent

The Intent Agent is a feature that users can enable to enhance AI efficiency. By providing descriptions for folders, the Intent Agent enables the AI to sift through the database and assess which information is most relevant to users. The AI uses these folder descriptions to identify and prioritize the folders most likely to contain relevant information. This feature transforms databases from simple file repositories into structured resources optimized for quick and accurate data retrieval.

How It Works

1. Folder Descriptions: Users provide brief descriptions for each uploaded folder, enabling the AI to better understand the folder's contents.

2. Intelligent Scanning: The AI first analyzes these folder descriptions before processing large volumes of data.

3. Relevance Assessment: The AI identifies which folders are most likely to contain relevant information.

4. Targeted Search: The AI focuses on exploring files within the most promising folders.

Benefits

• Improved Efficiency: Reduces time spent scanning irrelevant folders by narrowing the search to relevant areas.

• Optimized Workflows: Simplifies the process of locating the correct folder, especially as databases grow larger and more complex.

• Faster Results: Quickly identifies and retrieves the most relevant information, saving time.

• Resource Optimization: Conserves computational resources by prioritizing the processing of relevant data.

Outline for a Good Folder Description

Below is a sample folder description with three main parts: (a) short description, (b) details on the folder contents, and (c) keywords.

Example: Describing a Folder Filled with Administrative Forms

Folder Description:"This folder contains essential administrative and operational forms and documents for 'COMPANY A' including forms for employee onboarding, preference calculations, and quality management procedures. This folder consist of administrative and operational forms used to support various aspects of company operations, compliance, and employee management."Folder Description Breakdown:

short description	"Contains essential administrative and operational forms and documents for 'COMPANY A'..."
content details	"...including forms for employee onboarding, preference calculations, and quality management procedures..."
keywords	"...This folder consist of administrative and operational forms used to support various aspects of company operations, compliance, and employee management."

---

Bad Folder Descriptions Examples

1. Case 1: Too long. Descriptions that are redundant can reduce efficiency.

sample case	bad folder description	improved description
Describing a folder filled with administrative forms.	This folder contains a variety of forms and documents used for different administrative, operational, and compliance purposes. Here is a detailed summary of its contents: TR48-02: This form is used for documenting company property that is temporarily handed over to employees. It includes fields for item details, serial numbers, department, and signatures for both issuance and return. TR48-03: This form is an onboarding plan for new employees. It outlines the necessary training and introductions, including safety briefings, department tours, and specific job-related training. TR48-05: This document outlines the non-disclosure agreement terms between parties, including the return or destruction of confidential documents, rights to developments, and the governing laws. TR48-10: This form is used for supplier information, particularly regarding environmental management systems and compliance with various environmental regulations.	Contains essential administrative and operational forms and documents for 'COMPANY A,' including forms for employee onboarding, property management, non-disclosure agreements, supplier information, access rights and device requests, preference calculations, and quality management procedures.These documents support critical aspects of company operations, ensuring compliance and facilitating effective employee management.

1. Case 2: Too short. Descriptions that don't give enough context.

sample case	bad folder description	improved description
Describing a folder that includes all details to an internship and trainee program in the company	Folder for trainees	Contains essential documents related to the company's apprenticeship and training programs.This folder includes information on apprenticeship positions, trainer contact details, and guidelines for apprentices. Topics covered include working hour requirements, vacation entitlements, exam policies, and training report requirements, as well as details about additional training opportunities.It serves as a vital resource for both apprentices and trainers at 'COMPANY A,' supporting effective program management and development.

1. Case 3: Inconsistent keywords. Use words that are commonly used in the documents and widely understood by the company.

sample case	bad folder description	improved description
Describing a folder on the equipments and tools of a sales team.	Contains documents regarding information on the items used by the sales team	Contains essential documents for 'COMPANY A's' sales team (TeamB2B), including: An equipment list for the sales team's pilot case (Sales Kit). An overview of product sample cases (Sales Equipment Inventory). These documents detail the equipment, tools, and materials used for field sales activities, product demonstrations, and customer presentations, supporting the team's operational efficiency and effectiveness.

You can activate the Intent Agent in the Action Setting.

Step 10 - Save and share the bot

After configuring your bot, it's crucial to save your settings to ensure all customizations are preserved. To do this, simply click the Save button. This action will securely store all the changes you've made, safeguarding your bot's setup. In the Sharing Preferences you can also select your Bot to be Private or Public. If you turn this feature on, all your coworkers will be able to use this bot as well.