// the find
dair-ai/Prompt-Engineering-Guide
🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.
A documentation site covering prompt engineering techniques, from basic few-shot prompting to chain-of-thought, RAG, and agent patterns. It's aimed at developers and researchers who want a structured reference for how to get more out of LLMs, with accompanying papers and notebooks. Think of it as a living textbook rather than a library you import.
The breadth of technique coverage is genuinely useful — it doesn't just list CoT and stop, it includes less-obvious approaches like ART, Active-Prompt, and Directional Stimulus Prompting that most developers haven't encountered. The paper references are real academic citations, not just blog links, which matters if you need to understand why a technique works. The multilingual support (13 languages) is unusually thorough for a community-maintained doc site. The prompt hub section gives concrete, copy-pasteable examples organized by task type, which is more useful than abstract descriptions.
The last push was March 2026 but the model coverage section reads like 2023 — Phi-2 and Mistral 7B get dedicated pages while more recent and relevant models are missing or thin. The content is organized as documentation for a website (MDX files for Nextra), so contributing improvements requires knowing the site stack, not just writing markdown. There's no versioning or changelog for the guide content itself, so you can't tell when a specific technique page was last updated or whether the advice still applies to modern models. The sponsored-by placement at the top of the README is a mild credibility flag for what's otherwise positioning itself as neutral educational content.