LangGPT: Empowering everyone to become a prompt expert!🚀 Structured Prompt,Language of GPT, 结构化提示词,结构化Prompt
[2024.04.22 更新]:PromptShow 上线|轻松展示分享你的结构化提示词 欢迎访问 https://show.langgpt.ai/ 试用
[2024.04.13 更新]:提示词工程师必看系列:如何写好Prompt?
[2024.04.03 更新]:如何让 LLM 应用性能登峰造极
[2024.01.31 更新]:GPTs: LangGPT 代码诗人 自动化创建结构化提示词的简易版 GPTs, 持续更新中
[2024.01.14 更新]:年度分享:“AI 交流的艺术:Prompt Engineering 2023 年度视角”
[2023.12.20 更新]:【创业邦采访】欢迎来到提示词的美丽新世界
[2023.10.31 更新]:AI 超级个体见面会分享发言稿 和 PPT, 模板增加 tools 模块
[2023.9.03 更新]:BiliBili 视频教程(感谢 AIGCLINK)
[2023.8.31 更新]:Prompt 进阶 —— 提示链(Prompt Chain)和多提示词协同
[2023.7.30 更新]:系统论述:构建高性能 Prompt 之路——结构化 Prompt
The LangGPT project aims to facilitate the seamless creation of high-quality ChatGPT prompts for everyone by utilizing a structured, template-based methodology. It can be viewed as a programming language specifically crafted for designing prompts for large language models.
Current prompt design methods tend to offer only a handful of tips and principles, without a systematic and adaptable perspective. LangGPT transforms the prompt design process by incorporating templates, variables, and commands, enabling prompt creation to be as intuitive and straightforward as object-oriented programming. LangGPT sets the stage for the large-scale, efficient production of high-quality prompts.
With a solid grasp of LangGPT, you'll be able to quickly and effortlessly begin creating prompts for large language models in just a few minutes. 🚀 No need to save so many prompts anymore!
You can find out more details in the paper. LangGPT论文的中文版也已经上传。
更新内容 | 简介 | 更新日期 |
---|---|---|
LangGPT 论文 | LangGPT: Rethinking Structured Reusable Prompt Design Framework for LLMs from the Programming Language, 中文版 | 2024.2.26 |
GPTs: LangGPT 代码诗人 | 自动化创建结构化提示词的GPTs | 2024.1.31 |
视频 01-BiliBili 视频教程 | B 站 UP 主 AIGCLINK 制作的 LangGPT 视频教程!感谢 AIGCLINK! | 2023.9.03 |
文章 01-系统论述:构建高性能 Prompt 之路——结构化 Prompt | 关于结构化提示词看这篇就够了! | 2023.7.30 |
文章 02-Prompt 进阶 —— 提示链(Prompt Chain)和多提示词协同 | Prompt Chain 和 Prompt 协同,任务拆解等, LangGPT 社区联署联创文章. | 2023.8.31 |
知识库 01-LangGPT 提示词知识库 | LangGPT 提示词社区收集整理的提示词资料. | 2023.9.03 |
You can begin utilizing LangGPT for crafting potent prompts by simply continue the shared chat using GPT-4. This tool will empower you to craft, or adeptly transform traditional prompts into powerful LangGPT prompts.
You can find the corresponding Prompt in the LangGPT
folder. Note that the prompt is still in developing.
Here, we provide a small FitnessGPT
example to help you quickly get started with LangGPT. LangGPT offers prompt-writing templates, which you can use to rapidly create high-quality prompts.
# Role: FitnessGPT
## Profile
- Author: YZFly
- Version: 0.1
- Language: English
- Description: You are a highly renowned health and nutrition expert FitnessGPT. Take the following information about me and create a custom diet and exercise plan.
### Create custom diet and exercise plan
1. Take the following information about me
2. I am #Age years old, #Gender, #Height.
3. My current weight is #Currentweight.
4. My current medical conditions are #MedicalConditions.
5. I have food allergies to #FoodAllergies.
6. My primary fitness and health goals are #PrimaryFitnessHealthGoals.
7. I can commit to working out #HowManyDaysCanYouWorkoutEachWeek days per week.
8. I prefer and enjoy his type of workout #ExercisePreference.
9. I have a diet preference #DietPreference.
10. I want to have #HowManyMealsPerDay Meals and #HowManySnacksPerDay Snacks.
11. I dislike eating and cannot eat #ListFoodsYouDislike.
## Rules
1. Don't break character under any circumstance.
2. Avoid any superfluous pre and post descriptive text.
## Workflow
1. You will analysis the given the personal information.
2. Create a summary of my diet and exercise plan.
3. Create a detailed workout program for my exercise plan.
4. Create a detailed Meal Plan for my diet.
5. Create a detailed Grocery List for my diet that includes quantity of each item.
6. Include a list of 30 motivational quotes that will keep me inspired towards my goals.
## Initialization
As a/an <Role>, you must follow the <Rules>, you must talk to user in default <Language>,you must greet the user. Then introduce yourself and introduce the <Workflow>.
With the help of prompt above, you will create a Role named FitnessGPT, he/her will help you design wonderful personal diet and exercise plan.
Here are more LangGPT prompts. The examples
folder contains LangGPT prompt examples, including prompts and complete conversations with ChatGPT, to help you create wonderful prompt.
ChatGPT excels at role-playing. By providing role descriptions, role behaviors, and skills, it can produce actions that align well with the role.
Therefore, LangGPT designed the Role template to help ChatGPT better understand user intentions. The Role template is the core of LangGPT.
Here is the markdown Role template:
# Role: Your_Role_Name
## Profile
- Author: YZFly
- Version: 0.1
- Language: English or 中文 or Other language
- Description: Describe your role. Give an overview of the role's characteristics and skills
### Skill-1
1.skill description 1
2.skill description 2
### Skill-2
1.skill description 1
2.skill description 2
## Rules
1. Don't break character under any circumstance.
2. Don't talk nonsense and make up facts.
## Workflow
1. First, xxx
2. Then, xxx
3. Finally, xxx
## Tools
### browser
You have the tool `browser` with these functions:
- Issues a query to a search engine and displays the results.
- Opens the webpage with the given id, displaying it.
- Returns to the previous page and displays it.
- Scrolls up or down in the open webpage by the given amount.
- Opens the given URL and displays it.
- Stores a text span from an open webpage. Specifies a text span by a starting int `line_start` and an (inclusive) ending int `line_end`. To quote a single line, use `line_start` = `line_end`.
### python
When you send a message containing Python code to python, it will be executed in a
stateful Jupyter notebook environment. python will respond with the output of the execution or time out after 60.0
seconds. The drive at '/mnt/data' can be used to save and persist user files. Internet access for this session is disabled. Do not make external web requests or API calls as they will fail.
### dalle
Whenever a description of an image is given, use dalle to create the images and then summarize the prompts used to generate the images in plain text. If the user does not ask for a specific number of images, default to creating four captions to send to dalle that are written to be as diverse as possible.
### More Tools
## Initialization
As a/an <Role>, you must follow the <Rules>, you must talk to user in default <Language>,you must greet the user. Then introduce yourself and introduce the <Workflow>.
The Role template
primarily consists of four sections:
Profile
: The role's resume, including role description, characteristics, skills, and any other desired traits.Rules
: Rules the role must follow, usually involving actions they must take or avoid, such as "Never break role" and so on.Workflow
: The role's workflow, detailing the type of input users should provide and how the role should respond.Initialization
: Initializing the role according to the Role template's configuration, with most cases requiring only the default content.A role can be defined and configured using the four sections defined above.
Additionally, if you need to create complex prompts with commands, reminder, and other features, simply add the corresponding sections, as demonstrated in the advanced usage section.
Your_Role_Name
in Role: Your_Role_Name
with your desired role name.# Profile
section:
Language
as 中文
, English
, or any other language, using the target language for expression.Description
.### Skill
section. You can set multiple skills with bulleted descriptions for each skill.## Rules
: Add rules that the role must follow, typically covering required or prohibited actions, such as "Don't break role under any circumstance," etc.## Workflow
: Explain how the role should interact with users, the input users should provide, and how the role should respond.## Initialization
: The Role template sets up the role based on the template content, typically without modifications needed.As people continue to explore the capabilities of large models, LangGPT is still under development and refinement. Everyone is welcome to contribute to the LangGPT project, making it easier to use large models.
Variables offer significant versatility in prompt writing, simplifying the process of referencing role content, setting, and modifying role attributes.
This is an aspect that traditional prompt methods often find challenging to execute.
The Initialization
part of the Role template makes extensive use of variables:
As a/an <Role>, you must follow the <Rules>, you must talk to the user in the default <Language>, you must greet the user. Then introduce yourself and introduce the <Workflow>.
In LangGPT, variables are denoted by "<>". The variables here are:
<Role>
variable, representing the content of the entire Role.<Rules>
variable, representing the rules in the ## Rules
section.<Language>
variable, representing the value of the Language
field.Markdown's hierarchical structure allows ChatGPT to easily identify the content represented by variables:
Commands
make it easy to set some default actions, such as "/help" to provide help documentation, "/continue" to continue writing text
etc. which are all very useful commands.
## Commands
- Prefix: "/"
- Commands:
- help: This means that user do not know the commands usage. Please introduce yourself and the commands usage.
- continue: This means that your output was cut. Please continue where you left off.
Using a Reminder
can help alleviate ChatGPT's forgetting issue.
Add a Reminder
to the Role template:
## Reminder
1. 'Description: You will always remind yourself role settings and you output Reminder contents before responding to the user.'
2. 'Reminder: The user language is language (<language>), rules (<rules>).'
3. "<output>"
Use conditional statements just like in programming, with a template like:
If [situation1 happen], you will take [action1], else, you will take [action2]
Although LangGPT currently employs markdown language, any markup method capable of expressing hierarchical relationships, such as JSON or YAML, can also be utilized.
Maybe ChatGPT could assist in creating a conversion script?
The project is currently in its early and primitive stages, with a significant workload. We wholeheartedly welcome interested and skilled individuals to join and contribute to the project! 🆘
Task | Description | Status |
---|---|---|
Role Basic Template | Basic Prompt role design template, encompassing a majority of use cases | ✅ |
Documentation and Usage | Fundamentals of documentation, usage, and simple examples | ✅ |
Advanced Syntax Features | Develop more advanced syntax features alongside improvements in large model capabilities, such as longer context lengths, better long-term memory, and plugins | 📆 🆘 |
Prompt Chain | Collaboration between multi-role and prompt chains | 📆 🆘 |
Support for JSON/YAML | Support for JSON, YAML, and other markup formats to streamline development | 🔜 🆘 |
Role Advanced Template | Build upon the basic template by incorporating commands, environment settings, plugin functionality, network control, and other advanced features | 🔜 🆘 |
Examples | Supply more LangGPT template-based prompt examples and comprehensive conversation usage | ✅ |
Documentation | Enhance documentation and perfect usage | ✅ |
Website | Display documentation and examples for easy access | 📆 🆘 |
If you are not familiar with using GitHub, you can refer to: GitHub Minimal Contribution Guide: Issue and PR
If you used or referenced this project, please cite it in the following format:
@misc{wang2024langgpt,
title={LangGPT: Rethinking Structured Reusable Prompt Design Framework for LLMs from the Programming Language},
author={Ming Wang and Yuanzhong Liu and Xiaoming Zhang and Songlian Li and Yijie Huang and Chi Zhang and Daling Wang and Shi Feng and Jigang Li},
year={2024},
eprint={2402.16929},
archivePrefix={arXiv},
primaryClass={cs.SE}
}
Following projects provided great prompts, which inspired the creation of LangGPT: