[opensourcedex]

BotSharp is a free, open source ai interaction & interfaces project written in C# and released under Apache-2.0. It has 3,103 GitHub stars, 651 forks and 48 open issues, and was last pushed 29 hours ago. On this registry it ranks #76 of 135 tracked projects in AI Interaction & Interfaces, with 5 head-to-head comparisons available.

What is BotSharp?

BotSharp is an open-source AI multi-agent framework for .NET that connects large language models to existing business applications, and it is aimed at C# enterprise developers who want agent orchestration and conversation state to live inside their own information systems rather than in a separate stack.

What it is

BotSharp is an open source machine learning framework for AI bot platform builders, written in C# on .NET Core and therefore fully cross-platform. It covers natural language understanding, computer vision and audio processing, and it is built around a plug-in and pipeline flow execution design that decouples every component so that UI, model provider and pipeline stages can be swapped independently. An agent abstraction layer sits on top of that plumbing to manage complex enterprise application scenarios. The project is licensed under Apache-2.0, carries 3,103 stars and 651 forks on GitHub, and ships primarily through NuGet as BotSharp.Core.

The concrete problem it solves is integration cost. Connecting models to a business system normally means hand-writing model calls, prompt handling, conversation state and channel adapters into the application itself, once per project. BotSharp replaces that bespoke glue with a modular framework in the same language as the surrounding business logic, which matters because C# is a typed, enterprise-grade language widely used for information management systems and is easier to refactor at system scope. The README frames the goal directly: out-of-the-box machine learning algorithms let ordinary programmers develop AI applications faster and easier.

Key capabilities

  • Built-in multi-agent orchestration with conversation state management, where multiple agents with different responsibilities cooperate to complete complex tasks.
  • Multiple LLM planning approaches, so tasks ranging from simple to complex can be handled by different reasoning strategies.
  • Built-in RAG interfaces with memory based vector searching.
  • Provider-agnostic model support across ChatGPT 3.5, 4o, o1, Gemini 2, LLaMA 3, Claude Sonnet 3.5, DeepSeek V3 and HuggingFace.
  • A BotSharp UI written in SvelteKit for building, testing, evaluating and auditing LLM agents in one place.
  • An abstract standard Rich Content data structure integrated with popular message channels including Facebook Messenger, Slack and Telegram.
  • RESTful Open API plus WebSocket real-time communication, and built-in MCP integration with visual management of MCP servers that lets large models call tools, supporting mainstream services such as mcp.so.

Who uses it and how

  • Enterprise .NET teams that need to embed AI into existing information management systems without leaving the C# codebase.
  • Product teams building multi-agent workflows where separate agents hold distinct responsibilities and must coordinate on one task.
  • Teams deploying conversational agents across multiple channels, reusing the Rich Content structure over Facebook Messenger, Slack and Telegram.
  • Groups that need an evaluation and audit loop, using the SvelteKit UI to build, test, evaluate and audit agents in a single place.
  • Platform builders who want to swap LLM providers behind unified interfaces rather than committing to one vendor's models.

Getting started

Install the BotSharp.Core package from NuGet, or clone the repository and run the backend with dotnet run --project ./src/WebStarter/WebStarter.csproj -p SolutionName=BotSharp. The admin interface is a separate BotSharp UI project written in SvelteKit, and documentation is hosted at https://botsharp.readthedocs.io.

How it compares

No competing product is named in the supplied facts, so BotSharp stands alone in this registry as a .NET multi-agent framework. Its distinguishing position is provider neutrality: the same agent definitions can target ChatGPT, Gemini, LLaMA, Claude, DeepSeek or HuggingFace behind unified interfaces.

When to use it — and when not to

Self-hosting means running a .NET stack: the WebStarter backend plus a separate SvelteKit admin UI, with pipeline and plugin wiring to be configured by the operator. Teams with no .NET expertise, or those wanting a managed hosted service with no deployment work, should look elsewhere. Honest caveats: the README is an overview rather than an operations guide, its admin UI run steps are cut off in the excerpt, and 48 open issues remain to be worked through even though the repository is actively maintained.

project readme (upstream, from github) — read inline

The Open Source AI Agent Application Framework

Connect LLMs to your existing application focused on your business

Discord QQ群聊 Apache 2.0 NuGet build Documentation Status

"Conversation as a platform (CaaP) is the future, so it's perfect that we're already offering the whole toolkits to our .NET developers using the BotSharp AI BOT Platform Builder to build a CaaP. It opens up as much learning power as possible for your own robots and precisely control every step of the AI processing pipeline."

BotSharp is an open source machine learning framework for AI Bot platform builder. This project involves natural language understanding, computer vision and audio processing technologies, and aims to promote the development and application of intelligent robot assistants in information systems. Out-of-the-box machine learning algorithms allow ordinary programmers to develop artificial intelligence applications faster and easier.

It's written in C# running on .Net Core that is full cross-platform framework, the plug-in and pipeline flow execution design is adopted to completely decouple the plug-ins. C# is a enterprise grade programming language which is widely used to code business logic in information management related system. More friendly to corporate developers. BotSharp adopts machine learning algorithm in C# directly. That will facilitate the feature of the typed language C#, and be more easier when refactoring code in system scope.

BotSharp is in accordance with components principle strictly, decouples every part that is needed in the platform builder. So you can choose different UI/UX, or pick up a different LLM providers. They are all modularized based on unified interfaces. BotSharp provides an advanced Agent abstraction layer to efficiently manage complex application scenarios in enterprises, allowing enterprise developers to efficiently integrate AI into business systems.

Some Features

  • Built-in multi-agents and conversation with state management.
  • Support multiple LLM Planning approaches to handle different tasks from simple to complex.
  • Built-in RAG related interfaces, Memory based vector searching.
  • Support multiple AI platforms (ChatGPT 3.5/ 4o/ o1, Gemini 2, LLaMA 3, Claude Sonnet 3.5,DeepSeek V3, HuggingFace).
  • Allow multiple agents with different responsibilities cooperate to complete complex tasks.
  • Build, test, evaluate and audit your LLM agent in one place.
  • Build-in BotSharp UI written in SvelteKit.
  • Abstract standard Rich Content data structure. Integrate with popular message channels like Facebook Messenger, Slack and Telegram.
  • Provide RESTful Open API and WebSocket real-time communication.
  • Built-in MCP Integration: Visual management of MCP, enabling large model calls to Tools. Supports mainstream services like mcp.so

Quick Started

  1. Run backend service
 PS D:\> git clone https://github.com/dotnetcore/BotSharp
 PS D:\> cd BotSharp
 # For Windows
 PS D:\BotSharp\> dotnet run --project .\src\WebStarter\WebStarter.csproj -p SolutionName=BotSharp
 # For Linux
 $ dotnet run --project ./src/WebStarter/WebStarter.csproj -p SolutionName=BotSharp
  1. Run Admin UI project, reference to BotSharp UI.
PS D:\> git clone https://github.com/SciSharp/BotSharp-UI
PS D:\> cd BotSharp-UI
PS D:\> npm install
PS D:\> npm run dev

Access http://localhost:5015/

Online Demo with UI

Core Modules

The core module is mainly composed of abstraction and framework function implementation, combined with some common tools.

  • Plugin Loader
  • Hooking
  • Authentication
  • Agent Profile
  • Conversation & State
  • Routing & Planning
  • Templating
  • File Repository
  • Caching
  • Rich Content
  • LLM Provider

Plugins

BotSharp uses component design, the kernel is kept to a minimum, and business functions are implemented by external components. The modular design also allows contributors to better participate. Below are the built-in plugins:

Data Storages
LLMs
  • BotSharp.Plugin.Planner
  • BotSharp.Plugin.AzureOpenAI
  • BotSharp.Plugin.OpenAI
  • BotSharp.Plugin.AnthropicAI
  • BotSharp.Plugin.DeepSeekAI
  • BotSharp.Plugin.GoogleAI
  • BotSharp.Plugin.MetaAI
  • BotSharp.Plugin.HuggingFace
  • BotSharp.Plugin.LLamaSharp
  • BotSharp.Plugin.SemanticKernel
  • BotSharp.Plugin.SparkDesk
Messaging / Channel
  • BotSharp.OpenAPI
  • BotSharp.Plugin.ChatHub
  • BotSharp.Plugin.MetaMessenger
  • BotSharp.Plugin.Twilio
  • BotSharp.Plugin.TelegramBots
  • BotSharp.Plugin.WeChat
RAGs
  • BotSharp.Plugin.KnowledgeBase
  • BotSharp.Plugin.Qdrant
Visions
  • BotSharp.Plugin.PaddleSharp
Tools
  • BotSharp.Plugin.Dashboard
  • BotSharp.Plugin.RoutingSpeeder
  • BotSharp.Plugin.AudioHandler
  • BotSharp.Plugin.ChartHandler
  • BotSharp.Plugin.EmailHandler
  • BotSharp.Plugin.ExcelHandler
  • BotSharp.Plugin.FileHandler
  • BotSharp.Plugin.ImageHandler
  • BotSharp.Plugin.HttpHandler
  • BotSharp.Plugin.SqlDriver
  • BotSharp.Plugin.WebDriver
  • BotSharp.Plugin.PythonInterpreter
UIs
  • BotSharp.Plugin.ChatbotUI

Roadmap

  • A2A
  • MCP
  • Realtime
  • Computer Use
  • Browser Use
  • Database Assistant
  • Code Interpreter
  • Conversation Management
  • Multi-Agent Routing
  • Knowledge Base

FAQ

General Questions

What is BotSharp? BotSharp is an open-source .NET machine learning framework for building AI Bot platforms. It provides out-of-the-box components for natural language understanding, computer vision, and audio processing, enabling enterprise developers to integrate AI into business systems efficiently.

What .NET versions are supported? BotSharp runs on .NET Core, which is fully cross-platform. It supports .NET 6+ and works on Windows, Linux, and macOS.

Is BotSharp free to use? Yes, BotSharp is licensed under Apache 2.0, allowing free use in both personal and commercial projects.

LLM Provider Configuration

Which LLM providers are supported? BotSharp supports multiple AI platforms including ChatGPT (3.5/4o/o1), Gemini 2, LLaMA 3, Claude Sonnet 3.5, DeepSeek V3, HuggingFace, and more through its modular plugin system.

How do I configure a custom LLM provider? Create a plugin that implements the LLM provider interface, or use one of the built-in plugins like BotSharp.Plugin.OpenAI or BotSharp.Plugin.AnthropicAI. Configure the API keys in your appsettings.json.

Agent Development

How do I create a custom agent? Define an agent profile with its responsibilities, then implement the required hooks and plugins. BotSharp's advanced Agent abstraction layer handles the complex orchestration automatically.

Can multiple agents cooperate on a task? Yes, BotSharp supports multi-agent scenarios where agents with different responsibilities can cooperate to complete complex tasks. The routing and planning system handles agent coordination.

MCP Integration

What is MCP and how does BotSharp support it? MCP (Model Context Protocol) enables large model calls to external tools. BotSharp has built-in MCP integration with visual management capabilities, supporting mainstream services like mcp.so.

Deployment

How do I deploy BotSharp in production? Run the backend service using dotnet run and deploy it behind a reverse proxy (Nginx/IIS). The Admin UI (built with SvelteKit) can be deployed separately. See the documentation for detailed deployment guides.

Can I use BotSharp without the UI? Yes, BotSharp provides RESTful Open API and WebSocket real-time communication. You can integrate it directly with your own fro

readme truncated — read the full docs on github

Frequently asked questions

Is BotSharp free to use?

BotSharp is open source under the Apache-2.0 licence. There is no licence fee and no seat count — you can self-host it or, where the project offers one, pay a vendor for a managed version instead.

What does BotSharp do?

AI Multi-Agent Framework in .NET

What is BotSharp written in?

BotSharp is primarily written in C#. Its source is publicly available at https://github.com/SciSharp/BotSharp, and it has 3,103 GitHub stars.