Skip to content

Commit

Permalink
docs: Refactor README with structured feature descriptions
Browse files Browse the repository at this point in the history 
- Enhance README for both English and Chinese versions
- Restructure feature sections with numbered categories
- Add more detailed and systematic feature explanations
- Improve readability and visual hierarchy of feature descriptions
- Maintain consistent formatting across English and Chinese READMEs
  • Loading branch information
@pftom
pftom committed Feb 18, 2025
1 parent 893888d commit 5fcbe8a
Show file tree
Hide file tree
Showing 2 changed files with 104 additions and 74 deletions.
89 changes: 52 additions & 37 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -63,43 +63,58 @@ For core deployment tutorials, environment variable configuration, and FAQs, ple

View details in [CONTRIBUTING](./CONTRIBUTING.md).

## Key Features

1. **🧵 Multi-threaded Conversations**: Refly introduces an innovative multi-threaded conversation system that enables seamless switching between multiple independent discussion topics. This facilitates natural thought expansion and in-depth exploration, breaking free from traditional conversation constraints while building effective Agentic Workflows in human-AI collaboration.

<br />

2. **🤖 Multi-model Support**: Integrates 13+ leading AI models including DeepSeek R1, Claude 3.5 Sonnet, Google Gemini 2.0, and OpenAI O3-mini. Supports hybrid model usage, parallel processing, inter-model dialogue, and unified knowledge base integration.

<br />

3. **🎨 Multimodal Capabilities**: Handles diverse input formats including text, images, and files. Supports 7+ file formats (PDF, DOCX, RTF, TXT, MD, HTML, EPUB) and 7+ image formats (PNG, JPG, JPEG, BMP, GIF, SVG, WEBP). Enables batch selection and AI querying of files and images on the canvas.

<br />

4. **⚡️ AI-Powered Skills**: Leverages AI models to generate new content nodes from any selected canvas context. Features include AI web search, knowledge base search, smart question suggestions, and document generation. Integrates capabilities from Perplexity AI, Stanford Storm, and more into a unified workspace.

<br />

5. **🔍 Context Awareness**: Provides precise temporary knowledge base support for each conversation, ensuring accurate AI comprehension and responses. Similar to Cursor, users can flexibly select any canvas node or add references as context.

<br />

6. **📚 Knowledge Integration**: Enables importing diverse external resources and intelligently organizing them within the canvas. Build comprehensive knowledge systems and personalized thinking spaces, with semantic querying powered by RAG technology - creating a true second brain.

<br />

7. **✂️ Browser Extension Clipper**: Easily capture content from popular platforms like Github, Medium, Wikipedia, and Arxiv. Save entire web pages or specific quotes directly to your AI knowledge base, continuously building your second brain while boosting productivity.

<br />

8. **📌 Smart Citations**: Flexibly select and reference content from various sources including resources, documents, memos, and skill outputs. One-click contextual referencing enhances conversation accuracy and depth.

<br />

9. **✍️ AI Document Editor**: Features a powerful Markdown real-time editor with AI-assisted content selection, precise modifications, and refinements. Provides a Notion-like experience with advanced AI editing capabilities.

<br />
## ✨ Key Features

### `1` 🧵 Multi-threaded Conversation System
Built on an innovative multi-threaded architecture that enables parallel management of independent conversation contexts. Implements complex Agentic Workflows through efficient state management and context switching mechanisms, transcending traditional dialogue model limitations.

### `2` 🤖 Multi-model Integration Framework
- Integration with 13+ leading language models, including DeepSeek R1, Claude 3.5 Sonnet, Google Gemini 2.0, and OpenAI O3-mini
- Support for model hybrid scheduling and parallel processing
- Flexible model switching mechanism with unified conversation interface
- Multi-model knowledge base collaboration

### `3` 🎨 Multimodal Processing Capabilities
- File Format Support: 7+ formats including PDF, DOCX, RTF, TXT, MD, HTML, EPUB
- Image Processing: Support for mainstream formats including PNG, JPG, JPEG, BMP, GIF, SVG, WEBP
- Intelligent Batch Processing: Canvas multi-element selection and AI analysis

### `4` ⚡️ AI-Powered Skill System
Integrating advanced capabilities from Perplexity AI, Stanford Storm, and more:
- Intelligent web-wide search and information aggregation
- Vector database-based knowledge retrieval
- Smart query rewriting and recommendations
- AI-assisted document generation workflow

### `5` 🔍 Context Management System
- Precise temporary knowledge base construction
- Flexible node selection mechanism
- Multi-dimensional context correlation
- Cursor-like intelligent context understanding

### `6` 📚 Knowledge Base Engine
- Support for multi-source heterogeneous data import
- RAG-based semantic retrieval architecture
- Intelligent knowledge graph construction
- Personalized knowledge space management

### `7` ✂️ Intelligent Content Capture
- One-click content capture from mainstream platforms (Github, Medium, Wikipedia, Arxiv)
- Intelligent content parsing and structuring
- Automatic knowledge classification and tagging
- Deep knowledge base integration

### `8` 📌 Citation System
- Flexible multi-source content referencing
- Intelligent context correlation
- One-click citation generation
- Reference source tracking

### `9` ✍️ AI-Enhanced Editor
- Real-time Markdown rendering
- AI-assisted content optimization
- Intelligent content analysis
- Notion-like editing experience

## How to Use?

Expand Down
89 changes: 52 additions & 37 deletions README_CN.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -63,43 +63,58 @@ docker compose up -d

查看 [CONTRIBUTING](./CONTRIBUTING_CN.md) 了解更多信息。

## 关键特性

1. **🧵 多线程对话** ：Refly 采用创新的多线程对话机制，让您能够自由切换多个独立对话主题，实现更流畅自然的思维发散与深入探讨，有效突破传统对话的局限性，在人机协作过程中构建复杂且有效的 Agentic Workflow。

<br />

2. **🤖 多模型支持** ：支持 DeepSeek R1、Claude 3.5 Sonnet、Google Gemini 2.0、OpenAI O3-mini 等 13+ 主流模型，支持多模型混合和并行使用，支持多模型对话，支持多模型知识库整合。

<br />

3. **🎨 多模态支持** ：支持文本、图片、文件等多种模态输入，支持 7+ 文件格式，包括 PDF, DOCX, RTF, TXT, MD, HTML, EPUB 等，支持 7+ 图片格式，包括 PNG, JPG, JPEG, BMP, GIF, SVG, WEBP 等，支持在画布上对文件、图片进行批量选中操作提问 AI。

<br />

4. **⚡️ AI 驱动的技能** ：借助 AI 模型驱动，输入问题后，可灵活选取画布中的任一节点作为上下文，生成新的节点内容，包括 AI 全网搜索、AI 知识库搜索、AI 推荐提问和 AI 文档智能生成等，将 Perplexity AI、Standford Storm 等能力整合在一处工作空间。

<br />

5. **🔍 上下文** ：为每次对话提供精确的临时知识库支持，确保 AI 模型能准确理解和回应您的问题。与 Cursor 类似，您可以灵活选择画布中的任意节点或添加引用作为上下文。

<br />

6. **📚 知识库整合** ：支持导入多样化的外部资源，根据需求智能整合到画布中，构建完整的知识体系，打造个性化的思维空间，还能通过 RAG 等智能检索技术进行语义提问，是真正的第二大脑。

<br />

7. **✂️ 浏览器插件剪藏**：支持从 Github、Medium、Wikipedia、Arxiv 等主流网页中一键剪藏网页内容或者收藏金句片段，并加入到 AI 知识库中，持续构建第二大脑，提升创作效率。

<br />

8. **📌 引文** ：支持从各类资源、文档、备忘录或技能输出中灵活选取内容，一键添加为上下文引用，提升对话的准确性和深度。

<br />

9. **✍️ AI 文档编辑** ：除提供强大的 Markdown 实时编辑功能外，支持智能选中文档内容，根据您的需求进行 AI 辅助的精准修改和润色，提供给您一个类 Notion 的强大 AI 编辑器。

<br />
## 核心特性

### `1` 🧵 多线程对话系统
基于创新的多线程对话架构，支持并行管理多个独立会话上下文。通过高效的状态管理和上下文切换机制，实现复杂的 Agentic Workflow，突破传统对话模型的限制。

### `2` 🤖 多模型集成框架
- 集成 13+ 主流大语言模型，包括 DeepSeek R1、Claude 3.5 Sonnet、Google Gemini 2.0、OpenAI O3-mini 等
- 支持模型混合调度和并行处理
- 灵活的模型切换机制和统一的对话接口
- 多模型知识库协同

### `3` 🎨 多模态处理能力
- 文件格式支持：PDF、DOCX、RTF、TXT、MD、HTML、EPUB 等 7+ 种格式
- 图像处理：支持 PNG、JPG、JPEG、BMP、GIF、SVG、WEBP 等主流格式
- 智能批处理：支持画布多元素批量选择和 AI 分析

### `4` ⚡️ AI 驱动的技能系统
集成 Perplexity AI、Stanford Storm 等先进能力，提供：
- 智能全网搜索与信息聚合
- 基于向量数据库的知识检索
- 智能问题改写与推荐
- AI 辅助文档生成工作流

### `5` 🔍 上下文管理系统
- 精确的临时知识库构建
- 灵活的节点选择机制
- 多维度上下文关联
- 类 Cursor 的智能上下文理解

### `6` 📚 知识库引擎
- 支持多源异构数据导入
- 基于 RAG 的语义检索架构
- 智能知识图谱构建
- 个性化知识空间管理

### `7` ✂️ 智能内容采集
- 支持主流平台内容一键采集（Github、Medium、Wikipedia、Arxiv 等）
- 智能内容解析与结构化
- 自动知识分类与标签
- 深度知识库集成

### `8` 📌 引用系统
- 灵活的多源内容引用
- 智能上下文关联
- 一键引用生成
- 引用溯源支持

### `9` ✍️ AI 增强编辑器
- 实时 Markdown 渲染
- AI 辅助内容优化
- 智能内容分析
- 类 Notion 的编辑体验

## 如何使用 ？

Expand Down

0 comments on commit 5fcbe8a

Please sign in to comment.