promptslab/Awesome-Prompt-Engineering

What it solves

Working with large language models (LLMs) requires knowing how to phrase instructions, structure context, and chain reasoning steps to get useful results. Prompt engineering is the skill of doing this well, and the field is growing across research, industry, and academia. This project collects the best curated resources on the subject in one place, so people do not have to search fragmented sources to stay up to date or find the right technique.

The README also mentions "Context Engineering" as a related area — the practice of managing what information and how much of it goes into a model's context window, which has become increasingly important as models process longer inputs.

How it works

This is an "awesome list" style repository — a hand-curated, categorized index of external resources. It is organized in a table of contents with major sections:

• Papers: Research split into sub-topics like major surveys, prompt optimization, prompt compression, reasoning advances (like chain-of-thought and tree-of-thought), in-context learning, agentic prompting and multi-agent systems, multimodal prompting, structured output, prompt injection security, text-to-image, and text-to-music/audio generation.
• Tools and Code: Software for prompt management, LLM evaluation, agent frameworks, prompt optimization, red teaming and security, Model Context Protocol (MCP), vibe coding and AI coding assistants (CLI agents, IDEs, extensions, cloud platforms, open-source frameworks), plus other notable repositories.
• APIs, Datasets and Benchmarks, Models, and AI Content Detectors: Lists of external resources in each category.
• Books, Courses, Tutorials and Guides, Videos, and Communities: Learning material and places to connect with other practitioners.
• Autonomous Research & Self-Improving Agents: A section covering resources on agents that improve themselves.

The README also offers a "Start Here" path for newcomers, recommending a free short course on ChatGPT prompt engineering for developers, the DAIR.AI open-source prompting guide, official docs from OpenAI and Anthropic, an article on context engineering, and a large survey paper covering 58+ prompting techniques from 1,500+ papers.

The project lives online at promptslab.github.io, and contributors are welcome via pull requests. There is also a Discord community for discussion.

Who it's for

• Beginners who want a structured entry point into prompt engineering, with a recommended learning path, courses, and tutorials.
• Researchers and academics looking for a broad bibliography of papers on prompting techniques, surveys, security, multimodal generation, and reasoning.
• Practitioners and developers building LLM applications who need tools for prompt testing, evaluation, optimization, agent frameworks, and coding assistants.
• Security-minded folks interested in prompt injection defenses and red teaming resources.
• Anyone selling or teaching LLM skills — the list covers books, videos, communities, and even AI content detector resources.

Highlights

• Comprehensive and current: Covers everything from foundational papers (GPT-3 few-shot learning) through 2025 research on reasoning models, context engineering, and agentic systems.
• Strong reading path for beginners: The "Start Here" section explicitly sequences free courses, guides, official docs, and a major survey for newcomers.
• Deep security coverage: Dedicated sub-sections on prompt injection attacks, defenses, jailbreaking, and red teaming tools.
• Practical tooling catalog: Lists prompt management, evaluation, optimization, and agent frameworks, plus a wide range of AI coding tools from CLI agents to cloud platforms.
• Multimodal scope: Includes papers on text-to-image and text-to-audio generation prompting, plus visual prompting surveys.
• Community features: Pull requests are welcomed, Discord is available, and the project is Apache 2.0 licensed.