Skip to content

Commit 7b71801

Browse files
committed
docs: 更新 README 反映高级功能和图像生成已完成
## 变更内容 ### 核心特性更新 - ✅ 添加图像生成功能说明(T2I和I2I) - ✅ 添加高级测试模式功能说明 - ✅ 将这些功能从"预览版(Beta)"升级为正式功能 ### 高级功能章节重构 **之前状态**: "高级功能预览(Beta)" - 标注为开发中 **当前状态**: "高级功能" - 正式发布 #### 图像生成模式 - 文生图(T2I):文本提示词生成图像 - 图生图(I2I):基于本地图片变换优化 - 多模型支持:Gemini、Seedream等 - 模型参数:支持各模型特有参数配置(尺寸、风格等) - 预览与下载:实时预览生成结果,支持下载保存 #### 高级测试模式 - 上下文变量管理:自定义变量、批量替换 - 多轮会话测试:模拟真实对话场景 - 工具调用支持:Function Calling集成 - 灵活调试能力 ### 开发路线更新 - ✅ 高级模式:变量管理、上下文测试、工具调用 - ✅ 图像生成:文生图(T2I)和图生图(I2I)支持 - ❌ 移除"支持图片输入和多模态处理"(已完成) ## 文档同步 - README.md (中文版) - README_EN.md (英文版) - 添加图像模式文档链接 关联文档: docs/image-mode.md
1 parent ad15374 commit 7b71801

File tree

2 files changed

+41
-31
lines changed

2 files changed

+41
-31
lines changed

README.md

Lines changed: 21 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -46,20 +46,28 @@ Prompt Optimizer是一个强大的AI提示词优化工具,帮助你编写更
4646
- 📝 **双模式优化**:支持系统提示词优化和用户提示词优化,满足不同使用场景
4747
- 🔄 **对比测试**:支持原始提示词和优化后提示词的实时对比,直观展示优化效果
4848
- 🤖 **多模型集成**:支持OpenAI、Gemini、DeepSeek、智谱AI、SiliconFlow等主流AI模型
49+
- 🖼️ **图像生成**:支持文生图(T2I)和图生图(I2I),集成Gemini、Seedream等图像模型
50+
- 📊 **高级测试模式**:上下文变量管理、多轮会话测试、工具调用(Function Calling)支持
4951
- 🔒 **安全架构**:纯客户端处理,数据直接与AI服务商交互,不经过中间服务器
5052
- 📱 **多端支持**:同时提供Web应用、桌面应用、Chrome插件和Docker部署四种使用方式
5153
- 🔐 **访问控制**:支持密码保护功能,保障部署安全
5254
- 🧩 **MCP协议支持**:支持Model Context Protocol (MCP) 协议,可与Claude Desktop等MCP兼容应用集成
5355

54-
## 🚀 高级功能预览 (Beta)
56+
## 🚀 高级功能
5557

56-
> **预览环境**[https://prompt-dev.always200.com](https://prompt-dev.always200.com) | 欢迎体验新功能并反馈
58+
### 图像生成模式
59+
- 🖼️ **文生图(T2I)**:通过文本提示词生成图像
60+
- 🎨 **图生图(I2I)**:基于本地图片进行图像变换和优化
61+
- 📐 **灵活配置**:支持生成1-4张图片,自定义尺寸和参数
62+
- 🔌 **多模型支持**:集成Gemini、Seedream等主流图像生成模型
5763

58-
- 📊 **上下文变量管理**:自定义变量、多轮会话测试、变量替换预览
64+
### 高级测试模式
65+
- 📊 **上下文变量管理**:自定义变量、批量替换、变量预览
66+
- 💬 **多轮会话测试**:模拟真实对话场景,测试提示词在多轮交互中的表现
5967
- 🛠️ **工具调用支持**:Function Calling集成,支持OpenAI和Gemini工具调用
60-
- 🎯 **高级测试模式**更灵活的提示词测试和调试能力
68+
- 🎯 **灵活调试**更强大的提示词测试和调试能力
6169

62-
*注:高级功能正在开发完善中,未来版本将正式集成到主版本*
70+
详细使用说明请查看 [图像模式文档](docs/image-mode.md)
6371

6472
## 快速开始
6573

@@ -310,7 +318,7 @@ pnpm dev:fresh # 完整重置并重新启动开发环境
310318
- [x] 桌面应用发布
311319
- [x] MCP服务发布
312320
- [x] 高级模式:变量管理、上下文测试、工具调用
313-
- [ ] 支持图片输入和多模态处理
321+
- [x] 图像生成:文生图(T2I)和图生图(I2I)支持
314322
- [ ] 支持工作区/项目管理
315323
- [ ] 支持提示词收藏和模板管理
316324

@@ -361,25 +369,22 @@ pnpm dev:fresh # 完整重置并重新启动开发环境
361369
- 提供最完整、最稳定的功能体验
362370
- 从 [GitHub Releases](https://github.com/linshenkx/prompt-optimizer/releases) 下载
363371

364-
2. **使用Docker部署**(服务端方案)
365-
- Docker部署运行在服务端,没有浏览器跨域限制
366-
- 支持内网环境,数据不出内网
367-
- 请求流向:Docker容器→模型服务提供商
368-
369-
3. **使用自部署的API中转服务**(专业方案)
372+
2. **使用自部署的API中转服务**(专业方案)
370373
- 部署如OneAPI、NewAPI等开源API聚合/代理工具
371374
- 在设置中配置为自定义API端点
372375
- 请求流向:浏览器→中转服务→模型服务提供商
373376
- 完全控制安全策略和访问权限
374377

378+
**注意**:Web版(包括在线版、Vercel部署、Docker部署)都是纯前端应用,都会受到浏览器CORS限制。只有桌面版或使用API中转服务才能解决跨域问题。
379+
375380
#### Q4: 我已正确配置本地模型(如Ollama)的跨域策略,为什么使用在线版依然无法连接?
376381
**A**: 这是由浏览器的**混合内容(Mixed Content)安全策略**导致的。出于安全考虑,浏览器会阻止安全的HTTPS页面(如在线版)向不安全的HTTP地址(如您的本地Ollama服务)发送请求。
377382

378383
**解决方案**:
379-
为了绕过此限制,您需要让应用和API处于同一种协议下(例如,都是HTTP)。推荐以下几种方式
380-
1. **使用桌面版**:桌面应用没有浏览器限制,是连接本地模型最稳定可靠的方式
381-
2. **docker部署**:docker部署也是http
382-
3. **使用Chrome插件**:插件在某些情况下也可以绕过部分安全限制
384+
为了绕过此限制,您需要让应用和API处于同一种协议下(例如,都是HTTP)。推荐以下方式
385+
1. **使用桌面版**:桌面应用没有浏览器限制,是连接本地模型最稳定可靠的方式
386+
2. **使用Docker部署(HTTP)**:通过 `http://localhost:8081` 访问,与本地Ollama都是HTTP
387+
3. **使用Chrome插件**:插件在某些情况下也可以绕过部分安全限制
383388

384389
</details>
385390

README_EN.md

Lines changed: 20 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -46,20 +46,28 @@ Prompt Optimizer is a powerful AI prompt optimization tool that helps you write
4646
- 📝 **Dual Mode Optimization**: Support for both system prompt optimization and user prompt optimization to meet different usage scenarios
4747
- 🔄 **Comparison Testing**: Real-time comparison between original and optimized prompts for intuitive demonstration of optimization effects
4848
- 🤖 **Multi-model Integration**: Support for mainstream AI models including OpenAI, Gemini, DeepSeek, Zhipu AI, SiliconFlow, etc.
49+
- 🖼️ **Image Generation**: Support for Text-to-Image (T2I) and Image-to-Image (I2I) with models like Gemini, Seedream
50+
- 📊 **Advanced Testing Mode**: Context variable management, multi-turn conversation testing, Function Calling support
4951
- 🔒 **Secure Architecture**: Pure client-side processing with direct data interaction with AI service providers, bypassing intermediate servers
5052
- 📱 **Multi-platform Support**: Available as web application, desktop application, Chrome extension, and Docker deployment
5153
- 🔐 **Access Control**: Password protection feature for secure deployment
5254
- 🧩 **MCP Protocol Support**: Supports Model Context Protocol (MCP), enabling integration with MCP-compatible AI applications like Claude Desktop
5355

54-
## 🚀 Advanced Features Preview (Beta)
56+
## 🚀 Advanced Features
5557

56-
> **Preview Environment**: [https://prompt-dev.always200.com](https://prompt-dev.always200.com) | Experience new features and provide feedback
58+
### Image Generation Mode
59+
- 🖼️ **Text-to-Image (T2I)**: Generate images from text prompts
60+
- 🎨 **Image-to-Image (I2I)**: Transform and optimize images based on local files
61+
- 📐 **Flexible Configuration**: Generate 1-4 images with customizable dimensions and parameters
62+
- 🔌 **Multi-model Support**: Integrated with mainstream image generation models like Gemini, Seedream
5763

58-
- 📊 **Context Variable Management**: Custom variables, multi-turn conversation testing, variable replacement preview
64+
### Advanced Testing Mode
65+
- 📊 **Context Variable Management**: Custom variables, batch replacement, variable preview
66+
- 💬 **Multi-turn Conversation Testing**: Simulate real conversation scenarios to test prompt performance in multi-turn interactions
5967
- 🛠️ **Function Calling Support**: Function Calling integration with support for OpenAI and Gemini tool calling
60-
- 🎯 **Advanced Testing Mode**: More flexible prompt testing and debugging capabilities
68+
- 🎯 **Flexible Debugging**: Enhanced prompt testing and debugging capabilities
6169

62-
*Note: Advanced features are currently in development and will be officially integrated into the main version in future releases*
70+
For detailed usage instructions, please refer to the [Image Mode Documentation](docs/image-mode.md)
6371

6472
## Quick Start
6573

@@ -313,7 +321,7 @@ pnpm dev:fresh # Complete reset and restart development environment
313321
- [x] Desktop application release
314322
- [x] MCP service release
315323
- [x] Advanced mode: Variable management, context testing, function calling
316-
- [ ] Support for image input and multimodal processing
324+
- [x] Image generation: Text-to-Image (T2I) and Image-to-Image (I2I) support
317325
- [ ] Support for workspace/project management
318326
- [ ] Support for prompt favorites and template management
319327

@@ -363,25 +371,22 @@ For detailed project status, see [Project Status Document](docs/project-status.m
363371
- Provides the most complete and stable feature experience
364372
- Download from [GitHub Releases](https://github.com/linshenkx/prompt-optimizer/releases)
365373

366-
2. **Use Docker Deployment** (Server-side solution)
367-
- Docker deployment runs on the server side with no browser CORS restrictions
368-
- Supports internal network environments, data stays within your network
369-
- Request flow: Docker container → Model service provider
370-
371-
3. **Use Self-deployed API Proxy Service** (Professional solution)
374+
2. **Use Self-deployed API Proxy Service** (Professional solution)
372375
- Deploy open-source API aggregation/proxy tools like OneAPI, NewAPI
373376
- Configure as custom API endpoint in settings
374377
- Request flow: Browser → Proxy service → Model service provider
375378
- Full control over security policies and access permissions
376379

380+
**Note**: All web versions (including online version, Vercel deployment, Docker deployment) are pure frontend applications and subject to browser CORS restrictions. Only the desktop version or using an API proxy service can solve CORS issues.
381+
377382
#### Q4: I have correctly configured CORS policies for my local model (like Ollama), why can't I still connect using the online version?
378383
**A**: This is caused by the browser's **Mixed Content security policy**. For security reasons, browsers block secure HTTPS pages (like the online version) from sending requests to insecure HTTP addresses (like your local Ollama service).
379384

380385
**Solutions**:
381386
To bypass this limitation, you need to have the application and API under the same protocol (e.g., both HTTP). We recommend the following approaches:
382-
1. **Use the desktop version**: Desktop applications have no browser restrictions and are the most stable and reliable way to connect to local models.
383-
2. **Docker deployment**: Docker deployment also uses HTTP
384-
3. **Use Chrome extension**: Extensions can bypass some security restrictions in certain situations.
387+
1. **Use the desktop version**: Desktop applications have no browser restrictions and are the most stable and reliable way to connect to local models
388+
2. **Use Docker deployment (HTTP)**: Access via `http://localhost:8081`, both the app and local Ollama use HTTP
389+
3. **Use Chrome extension**: Extensions can bypass some security restrictions in certain situations
385390

386391
</details>
387392

0 commit comments

Comments
 (0)