Fully open-source, free, and commercially usable knowledge base project for intelligent customer service

Introduction

FastWiki is a high-performance knowledge base system built on the latest technology stack, designed for large-scale information retrieval and intelligent search. Leveraging Microsoft Semantic Kernel for deep learning and natural language processing, combined with .NET 8 and the MasaBlazor frontend framework, the backend uses .NET 8 + MasaFramework + SemanticKernel to deliver an efficient, user-friendly, and scalable intelligent vector search platform. Our goal is to provide an intelligent search solution that understands complex queries and helps users quickly and accurately obtain the information they need.

Tech Stack

• Frontend Framework: MasaBlazor with HttpClient for front-end/back-end separation
• Backend Framework: MasaFramework based on .NET 8, using MiniApis for webApi functionality and higher performance
• Vector Search Engine: PostgreSQL vector extension for optimized search performance
• Deep Learning & NLP: Microsoft Semantic Kernel for enhanced semantic understanding
• License: Apache-2.0, encouraging community contributions and use

Features

• Intelligent Search: Leverages Semantic Kernel’s deep learning and NLP to understand complex queries and deliver precise search results.
• High Performance: Vector search optimization via PostgreSQL’s vector extension ensures fast responses even with large data volumes.
• Modern Frontend: Built with MasaBlazor for responsive design and a user-friendly interface.
• Powerful Backend: Based on the latest .NET 8 and MasaFramework for code efficiency and maintainability.
• Open Source & Community-Driven: Licensed under Apache-2.0, encouraging developers and enterprises to use and contribute.

Partial Screenshot Introduction

Add a knowledge base:

Upload your documents:

Click to upload or drag documents here (currently only supports text files like md or txt; PDF and other formats will be supported later):

Below is the document template for uploading:

# Why Choose MASA Blazor?

## What is Blazor?

Blazor is a framework for building interactive client-side web UI with .NET:

- Create rich interactive UI using C# instead of JavaScript.
- Share server-side and client-side app logic written in .NET.
- Render UI as HTML and CSS for wide browser support, including mobile browsers.
- Integrate with modern hosting platforms such as Docker.

Using .NET for client-side web development provides the following advantages:
- Write code in C# instead of JavaScript.
- Leverage the existing .NET library ecosystem.
- Share app logic between server and client.
- Benefit from .NET’s performance, reliability, and security.
- Stay productive with Visual Studio on Windows, Linux, and macOS.
- Build on a stable, feature-rich, and easy-to-use set of common languages, frameworks, and tools.

## What is MASA Blazor?

MASA Blazor provides a standard base component library based on Material Design and BlazorComponent’s interactivity. It offers preset components for standard scenarios such as layout, dialog boxes, loading, global exception handling, and more. Starting from real-world usage scenarios, it aims to meet the needs of more users and use cases, shorten development cycles, improve development efficiency, and provide a complete web solution – MASA Blazor Pro.

## Why Choose MASA Blazor?

MASA Blazor follows the Material Design specification, with every component carefully designed for modularity, responsiveness, and excellent performance.

MASA Blazor is maintained and regularly upgraded by a professional full-time team, providing fast response times, diversified solutions, and enterprise-level support. It is already used by well-known companies, and the MASA team’s self-developed MASA Stack product line will continue to use it, ensuring project quality and the continuous addition of new components and features.

Besides offering many open-source projects for middle-platform scenarios, one of its core components – MASA Blazor – aims to be the most practical component library.

## What does Masa Blazor have to do with Token?

Token is a .NET programmer born in 2001, passionate about open source. Token often contributes open source code to Masa’s open source projects. Their connection feels destined; Masa Blazor is a great open source project.

## Advantages:

- Rich Components: Includes 1:1 restored Vuetify base components, many practical pre-built components, and deep .NET integrations including URL, breadcrumb, navigation three-link, advanced search, i18n, etc.
- UI Design Language: Modern design style with excellent multi-device user experience.
- Professional Examples: MASA Blazor Pro provides preset layouts for various common scenarios.
- Easy to Learn: Detailed getting-started documentation and free video tutorials (in progress).
- Active Community: Users encourage real-time interaction; contributions welcome to build the most open open-source community.
- Long-term Support: Maintained by a full-time team with enterprise-level support.

Click Next for data processing:

Here we provide direct splitting and Q&A splitting (not yet implemented).

We modified the custom processing parameters. This value affects document splitting. Proper splitting yields better responses; too large splitting may consume excessive tokens leading to insufficient funds. Then click Next:

Here you can see all files to be uploaded. The steps are: first upload files to the server, then add data to the backend and vectorize it. This process may take time depending on document content.

After upload, the list shows uploaded data. Click "View" to see all split document data:

Click "View Details" to see all information:

Click Applications -> Create Application:

Open the application, click "Select Knowledge Base", bind the newly added knowledge base to the current application, then click "Save Changes". This way, conversations will search the bound knowledge base. You can also modify application parameters such as opening remarks or role prompt definitions:

Then click Chat and enter the content from your knowledge base:

Q: "What is the relationship between Masa Blazor and Token?"

You can see the response effect from the knowledge base above. If you directly ask GPT, it wouldn't know who Token is! A knowledge base can compensate for AI’s shortcomings to some extent. For custom enterprise documentation, simply add all documents to the knowledge base and turn the application into a dialog — you can provide customers with a better document assistant experience, potentially replacing a lot of human customer service costs. If you have further customization needs, contact me via WeChat: hjl010426

Open Source

FastWiki is licensed under Apache-2.0, fully usable for commercial purposes without copyright disputes.

GitHub: https://github.com/239573049/fast-wiki
Gitee: https://gitee.com/hejiale010426/fast-wiki

FastWiki technical discussion group: