shuyu-labs/AntSK

An AI knowledge base/agent built with .Net 9, AntBlazor, Semantic Kernel, and Kernel Memory, supporting local offline AI large models. It can run offline without an internet connection. Supports Aspire for monitoring application data.

What it solves

AntSK is an enterprise-grade AI knowledge base and agent platform designed to fill the gap in the .NET ecosystem. It allows organizations to build private AI systems where data remains within their internal network, supporting local offline LLM execution and compatibility with domestic Chinese software (信创) standards for government and corporate use.

How it works

Built on .NET 9 and Blazor, the platform integrates Microsoft's Semantic Kernel for AI orchestration and Kernel Memory for RAG (Retrieval-Augmented Generation) capabilities. It supports a wide range of LLM providers, including cloud services (OpenAI, Azure OpenAI) and local engines (Ollama, LlamaFactory, llama.cpp). For knowledge management, it processes various document formats (PDF, Word, Excel, etc.), converts them into vectors using BGE-embedding models, and stores them in vector databases like Qdrant, Redis, or PostgreSQL (pgvector).

Who it’s for

  • .NET/C# Development Teams: Teams who want to build AI applications without switching to Python or Node.js.
  • Data-Sensitive Industries: Finance, healthcare, and government sectors requiring full offline deployment.
  • Enterprise IT: Organizations needing a turnkey "AI All-in-One Machine" (hardware + software) solution.

Highlights

  • Full .NET Stack: Developed with .NET 9 and Blazor Server for high stability and strong typing.
  • Offline-First: Supports completely air-gapped operation using local inference engines.
  • Extensible Plugin System: Offers both .NET DLL plugins and HTTP API plugins for integrating external business logic.
  • Broad Database Support: Compatible with multiple relational databases (including domestic Chinese ones like Dameng) and various vector stores.
  • Integrated Multimodal Capabilities: Includes built-in support for Stable Diffusion for image generation and OCR for text extraction.

Related

  • Project
    s2b-dev/smart-second-brain

    The ​​​  ​  ​  …...............

  • Project
    neo4j/neo4j-graphrag-python

    A Python library from Neo4j that lets you build Graph Retrieval‑Augmented Generation (GraphRAG) apps. It provides pipelines to turn text/PDFs into a Neo4j knowledge graph, utilities for vector indexing, hybrid retrievers that blend vector search with graph traversal, and a simple RAG API that plugs any supported LLM (OpenAI, Anthropic, Ollama, etc.). Install via `pip install neo4j-graphrag` (optional extras for your LLM), create a KG, index embeddings, then query with natural language.

  • Project
    supermemoryai/opensearch-ai

    A personalized AI search engine that learns from a user's web browsing habits to provide tailored search results and answers.

  • Project
    dominostars/playtranslate

    A real-time translation and language learning Android app that uses OCR to translate text from games, visual novels, and manga directly on the screen.