mcpField
A
AIME-BOX

AIME-BOX

github.com/AimeBoxFrameworks & SDKs61Updated 34w ago
Repository

About AIME-BOX

AIME-BOX is a multi-platform desktop chat client built with langchain and Electron. It supports multiple model providers, a local knowledge base, tool calls and several agents, with the goal of running fully offline where possible.

Key features

  1. 01Support for Ollama, OpenAI, Anthropic and others
  2. 02Local knowledge base retrieval using lancedb
  3. 03Tool calls such as web search and OCR
  4. 04Built-in data extraction, coding and translation agents
  5. 05Saved prompts and MCP connection

Use cases

  1. 01Chatting with local models without an internet connection
  2. 02Building a question-answering base over local files
  3. 03Extracting specified fields from files with an agent

MCP support

The client can connect to MCP servers and make their capabilities available for the model to call.

Details

From the project README

⚠️ 仓库已迁移

本项目已迁移到新的仓库地址：

👉 https://github.com/DarkNoah/aime-chat

Aime Box

一款基于 langchain + electron开发的多平台桌面端 Chat 客户端

支持本地知识库，tool调用，多个智能agent调用

目标尽量实现全离线本地可执行的智能agent

🖼 截图

image image image image

开始
  1. 添加模型提供商(如Ollama)
  2. 点击管理模型(启用需要的模型)
  3. 聊天页面即可对话聊天
  4. 下载一些本地模型 设置-本地模型管理(如国内调整HuggingfaceUrl为hf-mirror)
  • 推介下载
    • bge-m3 用于知识库检索
    • bge-reranker-large 用于知识库重排序
    • RapidOCR-json_v0.2.0 用于win环境下的ocr
    • matcha-icefall-zh-baker 用于语音生成
    • sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17 用于语音识别
    • silero_vad 用于分离说话场景
    • sherpa-onnx-pyannote-segmentation-3-0、3dspeaker_speech_eres2net_base_sv_zh-cn_3dspeaker_16k 用于说话人识别
    • rmbg-1.4 用于背景消除
🌟 特点
  • 多个模型供应商支持 ollama、openai、deepseek、anthropic等等
  • 多种工具调用支持 webloader、websearch、stt、tts、ocr等等
  • 本地知识库引擎，使用lancedb 进行本地知识库检索，支持image、docx、doc、text、url做为输入源
  • 多agent，目前有数据提取agent、代码助手、翻译助手
  • 提示词保存功能
  • MCP支持,可接入供模型调用
📝 Todo
  • mac平台兼容
  • 多语言支持
  • 文档对话功能类似于ChatDoc 支持docx、pdf、txt、url、mp4、音频文件等格式的对话
  • 浏览器browser-use
  • Tool自动创建工具，输入url或说明文档自动创建工具供模型调用
  • 更多的agents
  • a2a支持
  • 添加多个浏览器实例
Agent
  1. Extract 数据提取助手，用于对指定文件或文件夹下所有文件进行检索，根据用户给出的字段进行全文或部分抽取，markdown格式输出
  2. ScriptAssistant 对用户的任务生成一个python脚本和venv环境执行
🔌 技术栈

感谢以下开源项目的支持

如有缺漏请联系作者补充

🐞 Dev
# node 22
nvm install 22
nvm use 22

npm i -g node-gyp

npm install
npm run start
💼 Build
# window
npm run package
🌐 About

author: [email protected]

ChangeLog

CHANGELOG

License

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

© Aime Box

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.