Repo LLM Apps Limited info

dair-ai/Prompt-Engineering-Guide

Open-source guide with papers, lectures, notebooks and resources on prompt engineering, RAG and AI agents, available as a web site and runnable locally.

  • 78.9k GitHub stars
  • MDX
  • ⚖️ MIT
  • 🎯 Beginner
View on GitHub (opens in a new tab) Setup guide
pnpm i next react react-dom nextra nextra-theme-docs
dair-ai/Prompt-Engineering-Guide — repo preview

What it is

The Prompt Engineering Guide is a repository of guides, papers, lectures, references and tools for prompt engineering with large language models. It covers introductory topics, prompting techniques, applications, a prompt hub, model pages, and risks and misuses. A web version is hosted at promptingguide.ai, and the guide can also be run locally with Node and pnpm.

Who it's for

  • Developers designing prompting techniques that interface with LLMs and other tools
  • Researchers using prompt engineering to improve LLM performance on tasks like question answering and arithmetic reasoning
  • Learners who want a structured overview of prompting techniques, applications and tools
  • Contributors who want to add or check translations of the guide

Requirements

Requirements

  • Node >=18.0.0 (to run the guide locally)
  • pnpm (to run the guide locally)

Setup

  1. Install Node and pnpm

    Install Node >=18.0.0, then install pnpm if it is not already on your system (see https://pnpm.io/installation).

  2. Install dependencies

    Install the packages needed to run the guide locally.

    bash
    pnpm i next react react-dom nextra nextra-theme-docs

  3. Start the guide

    Boot the guide, then browse it at http://localhost:3000/.

    bash
    pnpm dev

Examples

Run the guide locally

bash
bash
pnpm i next react react-dom nextra nextra-theme-docs
pnpm dev

What it does: Installs the dependencies and starts the dev server; the guide is then available at http://localhost:3000/. Useful for checking a new translation.

Cite the guide

bibtex
bibtex
@article{Saravia_Prompt_Engineering_Guide_2022,
author = {Saravia, Elvis},
journal = {https://github.com/dair-ai/Prompt-Engineering-Guide},
month = {12},
title = {{Prompt Engineering Guide}},
year = {2022}
}

What it does: BibTeX entry the README asks users to use when citing the guide in work or research.

Pros & cons

Pros

  • Pro:Broad coverage: introduction, techniques (zero-shot, few-shot, chain-of-thought, RAG, ReAct and more), applications, prompt hub, models and risks
  • Pro:Includes a 1-hour video lecture, a notebook with code and slides
  • Pro:Available as a web version and can be run locally; MIT licensed
  • Pro:Supports 13 languages and accepts translations

Cons

  • Con:The README says the most up-to-date guides are on the website, so the repo content may lag
  • Con:Local setup is documented only for contributors checking changes, with no other usage instructions
  • Con:Some related courses are offered through a paid academy rather than in the repo