diff --git a/.vscode/settings.json b/.vscode/settings.json index ed09bc9..5a46197 100644 --- a/.vscode/settings.json +++ b/.vscode/settings.json @@ -30,6 +30,7 @@ "dylib", "esbuild", "esno", + "excludesfile", "execa", "fetti", "fettis", diff --git a/README.md b/README.md index 3de9e97..c20f32c 100644 --- a/README.md +++ b/README.md @@ -37,7 +37,7 @@ Use GPT-Runner to manage your AI presets, engage in AI-powered conversations wit - [Quick Start](#quick-start) - [The first way: CLI](#the-first-way-cli) - [The second way: VSCode Extension](#the-second-way--vscode-extension) -- [Documentation](#documentation) +- [GPT-Runner File](#gpt-runner-file) - [Road map](#road-map) - [What's New](#whats-new) - [FAQ](#faq) @@ -71,12 +71,16 @@ Use GPT-Runner to manage your AI presets, engage in AI-powered conversations wit ## Quick Start -> 1. Requirements NodeJS >= 16.15.0 -> - To check your NodeJS version, run `node -v` in your terminal. If you need to install or update NodeJS, visit [the official NodeJS website](https://nodejs.org/) for download and installation instructions. -> 2. Make sure you have an Open AI Key or a free ChatGPT account. +> Make sure you have an Open AI Key or a Anthropic Key. You can get it from [Open AI](https://platform.openai.com/) or [Anthropic](https://www.anthropic.com/). + ### The first way: CLI +> 1. Requirements NodeJS >= 16.15.0 +> - To check your NodeJS version, run `node -v` in your terminal. If you need to install or update NodeJS, visit [the official NodeJS website](https://nodejs.org/) for download and installation instructions. +> 2. First run the following command to download this package, which will take a long time, which is normal. + + ```bash cd npx gptr @@ -86,11 +90,15 @@ You can see the web interface in your browser at [http://localhost:3003](http:// ### The second way: VSCode Extension +> Requirements VSCode >= 1.72.0 + Install the [GPT-Runner VSCode Extension](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner) from the VSCode Marketplace. -## Documentation +## GPT-Runner File -Read the [documentation](https://gpt-runner.nicepkg.cn/) for installation instructions, usage guide, and more details on GPT-Runner's features. +For a detailed introduction about `.gpt-runner` directory, `*.gpt.md` files, and `gptr.config.json` files, please see here: + +[Introduction to GPT-Runner Files](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-en.gpt.md) ## Road map diff --git a/README_CN.md b/README_CN.md index 79113a0..ac517ad 100644 --- a/README_CN.md +++ b/README_CN.md @@ -16,11 +16,11 @@ [终端工具](https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-cli/) / [网页版](https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-web/) / [VSCode 扩展](https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-vscode/) / [反馈](https://github.com/Yidadaa/ChatGPT-Next-Web/issues) / [打赏开发者](https://github.com/nicepkg/gpt-runner/assets/35005637/98a4962a-8a2e-4177-8781-1e1ee886ecdc) -[cli-url]: https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-cli/ +[cli-url]: https://github.com/nicepkg/gpt-runner/blob/main/packages/gpt-runner-cli/README_CN.md [cli-image]: https://img.shields.io/badge/CLI-Node.js-green?logo=node.js -[web-url]: https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-web/ +[web-url]: https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-web/README_CN.md [web-image]: https://img.shields.io/badge/Web-React-blue?logo=react -[vscode-url]: https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-vscode/ +[vscode-url]: https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-vscode/README_CN.md [vscode-image]: https://img.shields.io/badge/VSCode-Extension-blue?logo=visualstudiocode @@ -33,7 +33,7 @@ - [快速开始](#快速开始) - [方式一：CLI](#方式一cli) - [方式二：VSCode 扩展](#方式二vscode-扩展) -- [文档](#文档) +- [GPT-Runner 文件](#gpt-runner-文件) - [开发计划](#开发计划) - [最新动态](#最新动态) - [常见问题](#常见问题) @@ -66,14 +66,16 @@ ## 快速开始 -> 1. 要求 NodeJS >= 16.15.0 -> - 要检查您的 NodeJS 版本，请在终端中运行 `node -v`。如果您需要安装或更新 NodeJS，请访问 [官方 NodeJS 网站](https://nodejs.org/) 以获取下载和安装说明。 -> -> 2. 确保你有一个 Open AI Key 或一个免费的 ChatGPT 账号。 -> 3. 确保你的命令终端能 ping 通 api.openai.com （如果你在中国大陆，你可能需要科学上网）。 +> 确保你有一个 Open AI Key 或 Anthropic Key，如果没有，请访问 [Open AI](https://platform.openai.com/) 或 [Anthropic](https://www.anthropic.com/product/) 申请。 +> 2. 确保你的命令终端能 ping 通 google.com （如果你在中国大陆，你可能需要科学上网）。 ### 方式一：CLI + +> 1. 要求 NodeJS >= 16.15.0 +> - 要检查您的 NodeJS 版本，请在终端中运行 `node -v`。如果您需要安装或更新 NodeJS，请访问 [官方 NodeJS 网站](https://nodejs.org/) 以获取下载和安装说明。 +> 2. 首次执行以下命令会下载这个包，时间会比较长，这是正常的。 + ```bash cd <你的项目路径> npx gptr @@ -83,11 +85,15 @@ npx gptr ### 方式二：VSCode 扩展 +> 要求 VSCode >= 1.72.0 + 从 VSCode Marketplace 安装 [GPT-Runner VSCode 扩展](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner)。 -## 文档 +## GPT-Runner 文件 -阅读 [文档](https://gpt-runner.nicepkg.cn/) 以获取安装说明、使用指南和有关 GPT-Runner 功能的更多详细信息。 +关于 `.gpt-runner` 目录和 `*.gpt.md` 文件和 `gptr.config.json` 文件详细介绍请看这里： + + [GPT-Runner 文件介绍](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-cn.gpt.md) ## 开发计划 diff --git a/docs/faq-cn.md b/docs/faq-cn.md index f9b3ad1..80cba6c 100644 --- a/docs/faq-cn.md +++ b/docs/faq-cn.md @@ -1 +1,2 @@ -faq cn +# 常见问题 + diff --git a/docs/faq-en.md b/docs/faq-en.md index dc5c971..4514b4c 100644 --- a/docs/faq-en.md +++ b/docs/faq-en.md @@ -1 +1 @@ -faq en +# FAQ diff --git a/docs/gpt-config-cn.gpt.md b/docs/gpt-config-cn.gpt.md new file mode 100644 index 0000000..12e6f65 --- /dev/null +++ b/docs/gpt-config-cn.gpt.md @@ -0,0 +1,304 @@ +# GPT-Runner 文件 + +
+目录
+ +- [GPT-Runner 文件](#gpt-runner-文件) + - [介绍](#介绍) + - [.gpt-runner 目录](#gpt-runner-目录) + - [gptr.config.ts/js/json 配置文件](#gptrconfigtsjsjson-配置文件) + - [xxx.gpt.md AI 预设文件](#xxxgptmd-ai-预设文件) + - [聊天模型配置](#聊天模型配置) + - [OpenAI](#openai) + - [Anthropic](#anthropic) + +
+ +## 介绍 + +1. 当你启动 GPT-Runner 时，它会读先读取项目级配置文件。 +2. 这不是必要的，但是当你想要覆盖一些全局配置，这很有用。 +3. 按优先级排序，优先读取最前面的配置文件。 + +```yml +/.gpt-runner/gptr.config.ts +/.gpt-runner/gptr.config.js +/.gpt-runner/gptr.config.json + +/gptr.config.ts +/gptr.config.js +/gptr.config.json +``` + +4. 接着 GPT-Runner 会深度检索当前文件夹下所有的 `*.gpt.md` 文件。 +5. 这个过程默认是跳过项目的 `.gitignore` 里的文件的。这很省时间。 +6. 你可以通过配置 `gptr.config.ts` 文件来改变检索范围。 +7. 每个 `*.gpt.md` 文件都会被解析成一个 AI 预设。 + +### .gpt-runner 目录 + +1. `/.gpt-runner/` 目录是一个特殊目录，即便你把它包含在 `.gitignore` 里，它也会被检索。这对希望 GPT-Runner 不入侵项目的人很有用。 +2. 你可以把 `gptr.config.json` 和 `*.gpt.md` 文件都放在这个目录里。 +3. 然后再在 `.gitignore` 里添加 `.gpt-runner` 。既可以同时保证项目的干净，又可以让 GPT-Runner 读取到配置文件。 +4. 如果你希望一劳永逸地 git ignore .gpt-runner 目录，你可以执行这句命令实现全局 git ignore: + +```bash +git config --global core.excludesfile '~/.gitignore_global' +echo '.gpt-runner' >> ~/.gitignore_global +``` + +## gptr.config.ts/js/json 配置文件 + +1. gpt.config.ts/js/json 是一个配置文件，它可以覆盖项目级别全局配置。 +2. 它的配置类型如下 + +```ts +export interface UserConfig { + /** + * 模型配置 + */ + model?: ModelConfig + + /** + * 深度检索包含的文件路径，支持 glob + * @default null + */ + includes?: string | RegExp | string[] | RegExp[] | ((source: string) => boolean) | null + + /** + * 深度检索排除的文件路径，支持 glob + * @default [ + "** /node_modules", + "** /.git", + "** /__pycache__", + "** /.Python", + "** /.DS_Store", + "** /.cache", + "** /.next", + "** /.nuxt", + "** /.out", + "** /dist", + "** /.serverless", + "** /.parcel-cache" + ] + */ + excludes?: string | RegExp | string[] | RegExp[] | ((source: string) => boolean) | null + + /** + * 是否跳过 .gitignore 里的文件 + * 建议开启，这样可以节省检索时间 + * @default true + */ + respectGitIgnore?: boolean +} + +export interface ModelConfig { + /** + * 模型类型 + */ + type?: 'openai' | 'anthropic' + + /** + * 模型名称 + */ + modelName?: string + + // ...更多配置请参考具体模型 +} +``` + +3. 你可以在 `gptr.config.ts` 里使用 `defineConfig` 函数来配置 `UserConfig` 类型的配置文件。你可以安装 `@nicepkg/gpt-runner` 包。 + +```bash +npm i @nicepkg/gpt-runner +``` + +4. 你可以新建一个 `gptr.config.ts` 文件，然后在里填上示例配置： + +```ts +import { defineConfig } from '@nicepkg/gpt-runner' + +export default defineConfig({ + model: { + type: 'openai', + modelName: 'gpt-3.5-turbo-16k', + temperature: 0.9, + }, +}) +``` + +5. 当然你也可以装上我们的 VSCode 插件，它会自动根据我们的 [JSON Schema](https://unpkg.com/@nicepkg/gpt-runner-shared@latest/dist/json-schema/user-config.json) 来提示你的配置文件。 +6. 这是一个极简的 `gptr.config.json` 的示例： + +```json +{ + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k" + } +} +``` + +7. 这是一个完整的 `gptr.config.json` 的示例： + +```json +{ + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": 2000, + "topP": 1, + "frequencyPenalty": 0, + "presencePenalty": 0 + }, + "includes": null, + "excludes": [ + "**/node_modules", + "**/.git", + "**/__pycache__", + "**/.Python", + "**/.DS_Store", + "**/.cache", + "**/.next", + "**/.nuxt", + "**/.out", + "**/dist", + "**/.serverless", + "**/.parcel-cache" + ], + "respectGitIgnore": true +} +``` + +## xxx.gpt.md AI 预设文件 + +1. `xxx.gpt.md` 文件是一个 AI 预设文件，一个文件代表一个 AI 角色。 +2. 比如一个 `uni-test.gpt.md` 是专门为这个项目写单元测试的，一个 `doc.gpt.md` 是专门为这个项目写文档的。它具备非常大的价值，而且这对团队成员来说是可以复用的。 +3. 为什么不是 `xxx.gpt.json` ？因为那样子的话， `System Prompt` 和 `User Prompt` 里的内容经常需要转义字符，这样子写起来很麻烦。 +4. `xxx.gpt.md` 易于书写，易于阅读，易于维护。 +5. 一个极简的 AI 预设文件如下： + +```md +\`\`\`json +{ + "title": "分类目录/AI角色名字" +} +\`\`\` + +# System Prompt + +你是一个编码高手，你擅长重构代码，请遵循SOLID和KISS和DRY原则，然后重构这段代码使其变得更好 + +``` +4. 一个完整的 AI 预设文件如下： + +```md +\`\`\`json +{ + "title": "分类目录/AI角色名字", + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": 2000, + "topP": 1, + "frequencyPenalty": 0, + "presencePenalty": 0 + } +} +\`\`\` + +# System Prompt + +你是一个编码高手，你擅长重构代码，请遵循SOLID和KISS和DRY原则，然后重构这段代码使其变得更好 + +# User Prompt + +当您使用此预设创建新聊天时，User Prompt 文本将自动填充聊天输入框，您可以在发送到人工智能机器人之前对其进行编辑 + +# 备注 + +这里可以写你的备注 + +`model` / `modelName` / `temperature` / `System Prompt` / `User Prompt` 都是**可选**参数，而且可定制参数还有非常多。 + +你还可以通过项目根目录下的 `gptr.config.json` 覆盖很多参数的默认值 +``` + +## 聊天模型配置 + +### OpenAI + +[官方请求参数文档](https://platform.openai.com/docs/api-reference/chat/create) + +```ts +export interface OpenaiModelConfig { + type: 'openai' + + /** + * 模型名称 + */ + modelName: string + + /** + * 温度 + */ + temperature?: number + + /** + * 最大回复 token 数量 + */ + maxTokens?: number + + /** + * 每个步骤考虑的 token 的总概率质量 + */ + topP?: number + + /** + * 根据频率惩罚重复的 token + */ + frequencyPenalty?: number + + /** + * Penalizes repeated tokens + */ + presencePenalty?: number +} +``` + +### Anthropic + +[官方请求参数文档](https://docs.anthropic.com/claude/reference/complete_post) + +```ts +export interface AnthropicModelConfig { + type: 'anthropic' + + /** + * 模型名称 + */ + modelName: string + + /** + * 温度 + */ + temperature?: number + + /** + * 最大回复 token 数量 + */ + maxTokens?: number + + /** + * 每个步骤考虑的 token 的总概率质量 + */ + topP?: number + + /** + * 仅从前 K 个选项中对每个后续选项进行采样 + */ + topK?: number +} +``` diff --git a/docs/gpt-config-en.gpt.md b/docs/gpt-config-en.gpt.md new file mode 100644 index 0000000..632efc1 --- /dev/null +++ b/docs/gpt-config-en.gpt.md @@ -0,0 +1,307 @@ +# GPT-Runner Files + +
+Table of Contents
+ +- [GPT-Runner Files](#gpt-runner-files) + - [Introduction](#introduction) + - [.gpt-runner Directory](#gpt-runner-directory) + - [gptr.config.ts/js/json Configuration Files](#gptrconfigtsjsjson-configuration-files) + - [xxx.gpt.md AI Preset Files](#xxxgptmd-ai-preset-files) + - [Chat Model Configuration](#chat-model-configuration) + - [OpenAI](#openai) + - [Anthropic](#anthropic) + +
+ +## Introduction + +1. When you start GPT-Runner, it first reads the project-level configuration file. +2. This isn't necessary, but it's useful when you want to override some global configurations. +3. Sorted by priority, it prefers to read the file at topmost. + +```yml +/.gpt-runner/gptr.config.ts +/.gpt-runner/gptr.config.js +/.gpt-runner/gptr.config.json + +/gptr.config.ts +/gptr.config.js +/gptr.config.json +``` + +4. Then GPT-Runner will deeply retrieve all `*.gpt.md` files under the current folder. +5. This process defaults to skipping the files in the project's `.gitignore` which saves time. +6. You can change the retrieval range by configuring the `gptr.config.ts` file. +7. Each `*.gpt.md` file is parsed into an AI preset. + +### .gpt-runner Directory + +1. `/.gpt-runner/` directory is a special directory. Even if you include it in `.gitignore`, it will be retrieved. This is useful for people who hope GPT-Runner doesn't intrude into the project. +2. You can put both `gptr.config.json` and `*.gpt.md` files in this directory. +3. Then add `.gpt-runner` in `.gitignore`. So you can keep the project clean and let GPT-Runner read the configuration files at the same time. +4. If you want to git ignore the `.gpt-runner` directory once and for all, you can execute this command to achieve global git ignore: + +```bash +git config --global core.excludesfile '~/.gitignore_global' + +echo '.gpt-runner' >> ~/.gitignore_global +``` + +## gptr.config.ts/js/json Configuration Files + +1. gpt.config.ts/js/json is a configuration file, it can override project-level global configurations. +2. Its configuration type is as follows + +```ts +export interface UserConfig { + /** + * Model configuration + */ + model?: ModelConfig + + /** + * Deep retrieval includes file paths, support glob + * @default null + */ + includes?: string | RegExp | string[] | RegExp[] | ((source: string) => boolean) | null + + /** + * Deep retrieval excludes file paths, support glob + * @default [ + "** /node_modules", + "** /.git", + "** /__pycache__", + "** /.Python", + "** /.DS_Store", + "** /.cache", + "** /.next", + "** /.nuxt", + "** /.out", + "** /dist", + "** /.serverless", + "** /.parcel-cache" + ] + */ + excludes?: string | RegExp | string[] | RegExp[] | ((source: string) => boolean) | null + + /** + * Skip the files in .gitignore + * Recommended to turn on, this can save retrieval time + * @default true + */ + respectGitIgnore?: boolean +} + +export interface ModelConfig { + /** + * Model type + */ + type?: 'openai' | 'anthropic' + + /** + * Model name + */ + modelName?: string + + // ...more configurations please refer to specific model +} +``` + +3. You can use `defineConfig` function in `gptr.config.ts` to configure `UserConfig` type configuration file. You can install `@nicepkg/gpt-runner` package. + +```bash +npm i @nicepkg/gpt-runner +``` + +4. You can create a new `gptr.config.ts` file, then fill in the sample configuration: + +```ts +import { defineConfig } from '@nicepkg/gpt-runner' + +export default defineConfig({ + model: { + type: 'openai', + modelName: 'gpt-3.5-turbo-16k', + temperature: 0.9, + }, +}) +``` + +5. Of course, you can also install our VSCode plugin, it will automatically prompt your configuration file based on our [JSON Schema](https://unpkg.com/@nicepkg/gpt-runner-shared@latest/dist/json-schema/user-config.json). +6. This is the simple example for `gptr.config.json`: + +```json +{ + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k" + } +} +``` + +7. This is the complete example for `gptr.config.json`: + +```json +{ + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": 2000, + "topP": 1, + "frequencyPenalty": 0, + "presencePenalty": 0 + }, + "includes": null, + "excludes": [ + "**/node_modules", + "**/.git", + "**/__pycache__", + "**/.Python", + "**/.DS_Store", + "**/.cache", + "**/.next", + "**/.nuxt", + "**/.out", + "**/dist", + "**/.serverless", + "**/.parcel-cache" + ], + "respectGitIgnore": true +} +``` + +## xxx.gpt.md AI Preset Files + +1. `xxx.gpt.md` files are AI preset files, each file represents an AI character. +2. For example, a `uni-test.gpt.md` is specifically for this project to write unit tests, and a `doc.gpt.md` is specifically for this project to write documentation. It has great value and can be reused by team members. +3. Why not `xxx.gpt.json`? Because in that case, the content within `System Prompt` and `User Prompt` often need to escape characters, which makes it very troublesome to write. +4. It's easy to write, read, and maintain `xxx.gpt.md`. +5. A minimalist AI preset file looks like this: + +```md +\`\`\`json +{ + "title": "Category/AI character name" +} +\`\`\` + +# System Prompt + +You're a coding master specializing in refactoring code. Please follow SOLID, KISS and DRY principles, and refactor this section of code to make it better. + +``` + +6. A complete AI preset file looks like this: + +```md +\`\`\`json +{ + "title": "Category/AI Character Name", + "model": { + "type": "openai", + "modelName": "gpt-3.5-turbo-16k", + "temperature": 0.9, + "maxTokens": 2000, + "topP": 1, + "frequencyPenalty": 0, + "presencePenalty": 0 + } +} +\`\`\` + + +# System Prompt + +You are a coding master, skilled at refactoring code. Please adhere to the SOLID, KISS and DRY principles, and refactor this code to make it better. + +# User Prompt + +When you use this preset to create a new chat, the User Prompt text will automatically fill in the chat input box. You can edit it before sending it to the AI robot. + +# Remark + +You can write your remarks here. + +`model` / `modelName` / `temperature` / `System Prompt` / `User Prompt` are all **optional** parameters, and there are many more to customize. + +You can also override many default parameter values through the `gptr.config.json` at the root directory of the project. +``` + +## Chat Model Configuration + +### OpenAI + +[Official Request Parameters Documentation](https://platform.openai.com/docs/api-reference/chat/create) + +```ts +export interface OpenaiModelConfig { + type: 'openai' + + /** + * Model name + */ + modelName: string + + /** + * Temperature + */ + temperature?: number + + /** + * Max reply token number + */ + maxTokens?: number + + /** + * Total probability mass of tokens per step + */ + topP?: number + + /** + * Penalize repeated tokens according to frequency + */ + frequencyPenalty?: number + + /** + * Penalizes repeated tokens + */ + presencePenalty?: number +} +``` + +### Anthropic + +[Official Request Parameters Documentation](https://docs.anthropic.com/claude/reference/complete_post) + +```ts +export interface AnthropicModelConfig { + type: 'anthropic' + + /** + * Model name + */ + modelName: string + + /** + * Temperature + */ + temperature?: number + + /** + * Max reply token number + */ + maxTokens?: number + + /** + * Total probability mass of tokens per step + */ + topP?: number + + /** + * Only sample subsequent choices from the top K options + */ + topK?: number +} +``` diff --git a/packages/README.md b/packages/README.md deleted file mode 100644 index 490b1cf..0000000 --- a/packages/README.md +++ /dev/null @@ -1 +0,0 @@ -# Gpt-runner Packages diff --git a/packages/gpt-runner-cli/README.md b/packages/gpt-runner-cli/README.md index 7f3a35d..93abdf6 100644 --- a/packages/gpt-runner-cli/README.md +++ b/packages/gpt-runner-cli/README.md @@ -23,6 +23,7 @@ GPT-Runner CLI 是一款功能强大的命令行工具，可帮助您管理 AI - [Installation](#installation) - [Quick Start](#quick-start) - [CLI Commands](#cli-commands) +- [GPT-Runner File](#gpt-runner-file) - [FAQ](#faq) - [Sponsor](#sponsor) - [Contributor](#contributor) @@ -43,7 +44,7 @@ GPT-Runner CLI 是一款功能强大的命令行工具，可帮助您管理 AI > 1. Requirements NodeJS >= 16.15.0 > - To check your NodeJS version, run `node -v` in your terminal. If you need to install or update NodeJS, visit [the official NodeJS website](https://nodejs.org/) for download and installation instructions. -> 2. Make sure you have an Open AI Key or a free ChatGPT account. +> 2. Make sure you have an Open AI Key or a Anthropic Key. You can get it from [Open AI](https://platform.openai.com/) or [Anthropic](https://www.anthropic.com/). > 3. Slow installation is normal, because the package is a bit big. To install the GPT-Runner CLI, use npm: @@ -106,8 +107,13 @@ gptr --port 8080 --config ./gptr.config.json --share --no-open --debug In this example, the GPT-Runner CLI will start a server available at port 8080, use the [gptr.config.json](https://github.com/nicepkg/gpt-runner/tree/main/docs/gptr.config.json) configuration file, share the server link, not automatically open in the browser, and run in debug mode. -For more detailed guidance, please refer to the project [documentation](https://gpt-runner.nicepkg.cn/). +## GPT-Runner File + +For a detailed introduction about `.gpt-runner` directory, `*.gpt.md` files, and `gptr.config.json` files, please see here: + +[Introduction to GPT-Runner Files](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-en.gpt.md) + ## FAQ [English > FAQ](https://github.com/nicepkg/gpt-runner/tree/main/docs/faq-en.md) diff --git a/packages/gpt-runner-cli/README_CN.md b/packages/gpt-runner-cli/README_CN.md index 1eac573..49a93b1 100644 --- a/packages/gpt-runner-cli/README_CN.md +++ b/packages/gpt-runner-cli/README_CN.md @@ -21,6 +21,7 @@ GPT-Runner CLI 是一款功能强大的命令行工具，可帮助您管理 AI - [安装](#安装) - [快速开始](#快速开始) - [CLI 命令](#cli-命令) +- [GPT-Runner 文件](#gpt-runner-文件) - [常见问题](#常见问题) - [赞助](#赞助) - [贡献者](#贡献者) @@ -40,8 +41,8 @@ GPT-Runner CLI 是一款功能强大的命令行工具，可帮助您管理 AI > 1. 要求 NodeJS >= 16.15.0 > - 要检查您的 NodeJS 版本，在终端运行 `node -v` 。如果需要安装或更新 NodeJS ，请访问[官方 NodeJS 网站](https://nodejs.org/)以获取下载和安装指南。 -> 2. 请确保您拥有一个 Open AI Key 或一个免费的 ChatGPT 帐户。 -> 3. 确保你的命令终端能 ping 通 api.openai.com （如果你在中国大陆，你可能需要科学上网）。 +> 2. 确保你有一个 Open AI Key 或 Anthropic Key，如果没有，请访问 [Open AI](https://platform.openai.com/) 或 [Anthropic](https://www.anthropic.com/product/) 申请。 +> 3. 确保你的命令终端能 ping 通 google.com （如果你在中国大陆，你可能需要科学上网）。 > 4. 安装速度较慢是正常的，因为软件包体积较大。 使用 npm 安装 GPT-Runner CLI : @@ -104,7 +105,11 @@ gptr --port 8080 --config ./gptr.config.json --share --no-open --debug 在这个例子中，GPT-Runner CLI 将在端口 8080 上启动一个服务器，使用 [gptr.config.json](https://github.com/nicepkg/gpt-runner/tree/main/docs/gptr.config.json) 的配置文件，分享服务器链接，不自动在浏览器中打开，并以调试模式运行。 -有关更详细的指导，可以参考项目[文档](https://gpt-runner.nicepkg.cn)。 +## GPT-Runner 文件 + +关于 `.gpt-runner` 目录和 `*.gpt.md` 文件和 `gptr.config.json` 文件详细介绍请看这里： + +[GPT-Runner 文件介绍](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-cn.gpt.md) ## 常见问题 diff --git a/packages/gpt-runner-core/README.md b/packages/gpt-runner-core/README.md index ee48928..949abca 100644 --- a/packages/gpt-runner-core/README.md +++ b/packages/gpt-runner-core/README.md @@ -1 +1,5 @@ -no... \ No newline at end of file +# GPT-Runner Core + +GPT-Runner Core is the core of GPT-Runner. + +It contains some parser of gpt runner config file and some llm connection utils. diff --git a/packages/gpt-runner-shared/LICENSE b/packages/gpt-runner-shared/LICENSE new file mode 100644 index 0000000..95ade51 --- /dev/null +++ b/packages/gpt-runner-shared/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2023-PRESENT Jinming Yang + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/packages/gpt-runner-shared/README.md b/packages/gpt-runner-shared/README.md index ee48928..22ec3fc 100644 --- a/packages/gpt-runner-shared/README.md +++ b/packages/gpt-runner-shared/README.md @@ -1 +1,7 @@ -no... \ No newline at end of file +# GPT-Runner Shared + +GPT-Runner Shared is the shared code of GPT-Runner. + +It contains `browser` and `server` and `common` code for reused in all GPT-Runner packages. + +The `common` code can be used in both client side and server side. diff --git a/packages/gpt-runner-shared/src/common/zod/config/user.config.ts b/packages/gpt-runner-shared/src/common/zod/config/user.config.ts index d4e7688..1f28edc 100644 --- a/packages/gpt-runner-shared/src/common/zod/config/user.config.ts +++ b/packages/gpt-runner-shared/src/common/zod/config/user.config.ts @@ -38,6 +38,7 @@ export const UserConfigSchema = z.object({ export const UserConfigForUserSchema = UserConfigSchema.omit({ rootPath: true, + exts: true, }) satisfies z.ZodType export const SingleChatMessageSchema = z.object({ diff --git a/packages/gpt-runner-vscode/README.md b/packages/gpt-runner-vscode/README.md index ee48928..dcf9e61 100644 --- a/packages/gpt-runner-vscode/README.md +++ b/packages/gpt-runner-vscode/README.md @@ -1 +1,90 @@ -no... \ No newline at end of file +
+icon + +

GPT-Runner VSCode

+ +[简体中文](https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-cli/README_CN.md) / English + +[![VSCode](https://img.shields.io/badge/VSCode-Extension-blue?logo=visualstudiocode)](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner) [![License](https://img.shields.io/github/license/nicepkg/gpt-runner)](https://github.com/nicepkg/gpt-runner/blob/main/LICENSE) +![GitHub stars](https://img.shields.io/github/stars/nicepkg/gpt-runner?style=social) + +The GPT-Runner VSCode extension is the result of integrating GPT-Runner and VSCode. + +This extension helps you manage AI presets and drive conversations with code using AI, thereby significantly boosting your development efficiency. + +
+ +
+Contents
+ +- [Features](#features) +- [Installation](#installation) +- [Quick Start](#quick-start) +- [GPT-Runner File](#gpt-runner-file) +- [Common Questions](#common-questions) +- [Sponsor](#sponsor) +- [Contributors](#contributors) +- [License](#license) + +
+ +## Features + +- **Real-time context:** It can observe the files you have edited and the text you have selected to answer your questions. +- **Multiple chat window instances:** You can not only open a chat in the sidebar, but you can also open a chat page in the editor. +- **Responsive UI:** It adjusts its layout automatically based on the current window width. +- **Shareable LAN links:** Click the `🚀 GPT-Runner` at the bottom to open it in the browser. +- **Conversation with code files:** Select files or folders to chat with AI in real-time. +- **Manage your AI presets:** Manage your AI presets, it's like the local Storybook for AI presets. +- **Custom AI Parameters:** Flexible control over the configuration of AI models. +- **Support for third-party LLMs:** High compatibility and adaptability. +- **Privacy-First:** Local data storage to protect your privacy. +- **Localization:** Support for multiple languages. + +## Installation + +> 1. Requires VSCode >= 1.72.0 +> 2. Make sure you have an Open AI Key or Anthropic Key. If you do not have, please visit [Open AI](https://platform.openai.com/) or [Anthropic](https://www.anthropic.com/product/) to apply. + +You can search `GPT Runner` in the VSCode Extension Marketplace for installation. + +Or, click here to install [GPT-Runner VSCode Extension](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner). + +## Quick Start + +1. Open a project and click on the chat icon on the sidebar. +2. Click to confirm the initialization of the `.gpt.md` file. +3. Click settings in the top right corner and fill in your `Open AI Key` or `Anthropic Key`. +4. Click preset, then click `+` to create a new chat. +5. You are free to explore the functions within. + +> 1. Note the chat icon in the top right corner, which is also clickable +> 2. Note the status bar at the bottom, `🚀 GPT-Runner` is also clickable + +## GPT-Runner File + +For a detailed introduction about `.gpt-runner` directory, `*.gpt.md` files, and `gptr.config.json` files, please see here: + + [Introduction to GPT-Runner Files](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-en.gpt.md) + +## Common Questions + +[English > Common Questions](https://github.com/nicepkg/gpt-runner/tree/main/docs/faq-en.md) + +## Sponsor + +Waiting for your sponsorship... + +## Contributors + +You can check out our [Contribution Guidelines](https://github.com/nicepkg/gpt-runner/tree/main/CONTRIBUTING.md) + +This project exists thanks to all the people who contribute: + + + + + +## License + +[MIT](https://github.com/nicepkg/gpt-runner/tree/main/LICENSE) License © 2023-PRESENT [Jinming Yang](https://github.com/2214962083) diff --git a/packages/gpt-runner-vscode/README_CN.md b/packages/gpt-runner-vscode/README_CN.md new file mode 100644 index 0000000..d803cc8 --- /dev/null +++ b/packages/gpt-runner-vscode/README_CN.md @@ -0,0 +1,92 @@ +
+icon + +

GPT-Runner VSCode

+ +[English](https://github.com/nicepkg/gpt-runner/tree/main/packages/gpt-runner-cli/README.md) / 简体中文 + + +[![VSCode](https://img.shields.io/badge/VSCode-Extension-blue?logo=visualstudiocode)](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner) [![License](https://img.shields.io/github/license/nicepkg/gpt-runner)](https://github.com/nicepkg/gpt-runner/blob/main/LICENSE) +![GitHub stars](https://img.shields.io/github/stars/nicepkg/gpt-runner?style=social) + +GPT-Runner VSCode 扩展是 GPT-Runner 跟 VSCode 深度结合的结果。 + +可帮助您管理 AI 预设并与代码进行 AI 驱动的对话，从而显着提高您的开发效率。 + +
+ +
+目录
+ +- [特性](#特性) +- [安装](#安装) +- [快速开始](#快速开始) +- [GPT-Runner 文件](#gpt-runner-文件) +- [常见问题](#常见问题) +- [赞助](#赞助) +- [贡献者](#贡献者) +- [许可证](#许可证) + +
+ +## 特性 + +- **实时上下文:** 它可以观察你已打开编辑的文件和你选中的文字回答你的问题。 +- **多个聊天窗口实例:** 你不但可以在侧边栏打开聊天，还可以在编辑器里打开聊天页。 +- **响应式 ui:** 它会根据当前窗口宽度自动调整自身布局。 +- **可共享的局域网链接:** 点击底部 `🚀 GPT-Runner` 可以在浏览器打开。 +- **与代码文件对话：** 选择文件或文件夹与 AI 实时对话。 +- **管理你的 AI 预设：** 管理你的 AI 预设，它就像 AI 预设的本地 Storybook。 +- **自定义 AI 参数：** 灵活控制 AI 模型的配置。 +- **支持第三方 LLM：** 具有高度兼容性和适应性。 +- **隐私优先：** 本地数据存储保护您的隐私。 +- **国际化：** 支持多种语言。 + +## 安装 + +> 1. 要求 VSCode >= 1.72.0 +> 2. 确保你有一个 Open AI Key 或 Anthropic Key，如果没有，请访问 [Open AI](https://platform.openai.com/) 或 [Anthropic](https://www.anthropic.com/product/) 申请。 + + +你可以在 VSCode 扩展市场搜索 `GPT Runner` 进行安装。 + +或者点击这里安装 [GPT-Runner VSCode 扩展](https://marketplace.visualstudio.com/items?itemName=nicepkg.gpt-runner)。 + +## 快速开始 + +1. 打开一个项目，点击侧边聊天 icon。 +2. 点击确认初始化 `.gpt.md` 文件。 +3. 点击右上角设置，填入你的 `Open AI Key` 或 `Anthropic Key`。 +4. 点击预设，点击 `+` 新建聊天。 +5. 你可以自由探索里面的功能。 + +> 1. 注意你的右上角有个聊天 icon，那个也能点击 +> 2. 注意你的底部状态栏，有个 `🚀 GPT-Runner` 也能点击 + +## GPT-Runner 文件 + +关于 `.gpt-runner` 目录和 `*.gpt.md` 文件和 `gptr.config.json` 文件详细介绍请看这里： + + [GPT-Runner 文件介绍](https://github.com/nicepkg/gpt-runner/blob/main/docs/gpt-config-cn.gpt.md) + +## 常见问题 + +[简体中文 > 常见问题](https://github.com/nicepkg/gpt-runner/tree/main/docs/faq-cn.md) + +## 赞助 + +等待你的赞助... + +## 贡献者 + +你可以查看我们的[贡献指南](https://github.com/nicepkg/gpt-runner/tree/main/CONTRIBUTING.md) + +这个项目得以存在，要感谢所有贡献者： + + + + + +## 许可证 + +[MIT](https://github.com/nicepkg/gpt-runner/tree/main/LICENSE) 许可证 © 2023-PRESENT [Jinming Yang](https://github.com/2214962083) diff --git a/packages/gpt-runner-web/LICENSE b/packages/gpt-runner-web/LICENSE new file mode 100644 index 0000000..95ade51 --- /dev/null +++ b/packages/gpt-runner-web/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2023-PRESENT Jinming Yang + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/packages/gpt-runner-web/README.md b/packages/gpt-runner-web/README.md index ee48928..2b697ed 100644 --- a/packages/gpt-runner-web/README.md +++ b/packages/gpt-runner-web/README.md @@ -1 +1,9 @@ -no... \ No newline at end of file +# Gpt-runner Web + +Gpt-runner Web is the web interface of GPT-Runner. + +It contains both server side and client side code. + +Server side is power by express. + +Client side is power by react + vite. diff --git a/packages/gpt-runner/README.md b/packages/gpt-runner/README.md index 4560e0c..fd2b713 100644 --- a/packages/gpt-runner/README.md +++ b/packages/gpt-runner/README.md @@ -1,61 +1,5 @@ # GPT-Runner -The GPT-Runner package is a powerful and versatile AI-driven tool designed to help developers improve their workflow by integrating AI-powered robot presets into their development environment. This package is a part of the larger GPT-Runner project and serves as the core engine behind its functioning. +It contains `defineConfig` function and `UserConfig` type. -## Overview - -GPT-Runner enables developers to create and manage AI-powered robot presets through the use of `xxx.gpt.md` files within their projects. These files contain customizable parameters such as system prompts, user prompts, model names, and temperature, allowing developers to fine-tune the AI settings to suit their specific needs. - -By integrating GPT-Runner into their development workflow, developers can enjoy features such as: - -- AI Robot Preset-based Chats for efficient team communication and collaboration. -- Seamless integration with any IDE through the GPT-Runner CLI and IDE extensions. -- Customizable AI parameters for enhanced control and adaptability. - -## Installation - -To install and utilize the GPT-Runner package in your project, follow these steps: - -1. Ensure that you have Node.js (version 16.15.0 or higher) installed on your system. -2. Add the GPT-Runner package as a dependency in your project by running the following command: - - ```sh - npm install @nicepkg/gpt-runner - ``` - -3. Import the GPT-Runner package in your project and utilize its features according to your requirements. - -## Usage - -Define your project configuration using the `defineConfig` function, providing it with a configuration object containing your project's preferences: - -```javascript -import { defineConfig } from '@nicepkg/gpt-runner'; - -const config = defineConfig({ - rootPath: 'path/to/your/project/root', - model: { - type: 'openai', - modelName: 'gpt-4', - secrets: { - apiKey: 'your_api_key' - } - } -}); -``` - -### Other Functions and Utilities - -GPT-Runner exposes various other functions and utilities from the shared `@nicepkg/gpt-runner-shared` package, which you can utilize in your project as required. Explore the GPT-Runner package source code and documentation for more information on the available functions and their use cases. - -## Troubleshooting - -If you encounter any issues or need more information on the GPT-Runner package, refer to the [FAQ](https://github.com/nicepkg/gpt-runner/blob/main/website/faq-en.md) sections in the project documentation. If you can't find a solution there, feel free to open an [issue](https://github.com/nicepkg/gpt-runner/issues) on the project's GitHub page. - -## Contributing - -Contributions to the GPT-Runner package are welcome! Check out the [contributing guideline](https://github.com/nicepkg/gpt-runner/blob/main/CONTRIBUTING.md) to get started. - -## License - -The GPT-Runner package is provided under the [MIT License](https://github.com/nicepkg/gpt-runner/blob/main/LICENSE). +It's helpful to config `gptr.config.ts` file with type checking. diff --git a/packages/gpt-runner/src/index.ts b/packages/gpt-runner/src/index.ts index 6662fc4..c84c910 100644 --- a/packages/gpt-runner/src/index.ts +++ b/packages/gpt-runner/src/index.ts @@ -6,7 +6,7 @@ export * from '@nicepkg/gpt-runner-shared/node' type GetStaticValueFromChatModelType = T extends ChatModelType ? `${T}` : never export type UserConfig = { - [Key in keyof IUserConfig]: Key extends 'model' ? Omit, 'type'> & { + [Key in keyof Omit]: Key extends 'model' ? Omit, 'type'> & { type: GetStaticValueFromChatModelType['type']> } : IUserConfig[Key] } diff --git a/website/README.md b/website/README.md index 57ee3a5..70ecff9 100644 --- a/website/README.md +++ b/website/README.md @@ -1 +1,5 @@ -...no +# GPT-Runner Website + +Copy from docusaurus official code. + +Wait for work... （´・ω・`）