This curated guide brings together 28 vetted Prompt Engineering resources, including 26 with structured technical analysis and 0 with practitioner field notes. Primary focus areas include AI Agent Development, AI Agent Workflow, AI Architecture Blog Post, AI Coding Guidelines, AI Coding Tool. Primary sources include github.com, softwaredoug.com, blog.kunchenguid.com, cookbook.openai.com, gitingest.com.
| Resource | Category & Format |
|---|
| Key Takeaway & Notes |
|---|
| Don’t classify. Hallucinate!softwaredoug.com | LLM Engineering Pattern / blog post with code examples / free online | The classic approach of using structured outputs with a giant Literal of 500+ valid categories is expensive and hits token limits |
|---|---|---|
| Design the workwork-guide.openai.chatgpt.site | AI Productivity Guide / interactive guide with video walkthroughs and downloadable markdown / free online | Six-stage framework: mindset, workflow, skills, automation, and project management for structuring AI-assisted work |
| Repo Promptrepoprompt.com | AI Coding Tool / macOS native application / Paid with Pro upgrade tier | Advanced file selection with token estimation to send only relevant code to LLMs, reducing cost and noise |
| Prompting Claude Fable 5.1platform.claude.com | Technical Documentation / Technical documentation / Free online | Provides specific prompting patterns to correct behavioral differences in Claude Fable 5.1 |
| Your AGENTS.md is a Neural Netblog.kunchenguid.com | AI Agent Development / long-form blog post / free online | Project-level AGENTS.md files should be treated as neural net weights with a fixed token budget, trained on actual session transcripts rather than edited transactionally. |
| peculiar-ragdoll/Qwen-Sharp-Chat-Templates · Hugging Facehuggingface.co | LLM Prompt Template / Open source repository with documentation / Free, open source on Hugging Face | Eleven-line terseness prompt that cuts filler without sacrificing correctness, making models more intelligent per token |
| claude-cookbooks/cost_optimization/cost_optimization.ipynb at main · anthropics/claude-cookbooksgithub.com | Technical Notebook / Jupyter notebook / Free, open source on GitHub | Official Anthropic resource for reducing Claude API costs |
| GitHub - staltz/microskills: Modular prompts I use every daygithub.com | Open Source Prompt Library / GitHub Repository / open source | Focuses on handwritten, non-verbose instructions to prevent context pollution in LLMs. |
| agent-scripts/skills/codex-first/SKILL.md at main · steipete/agent-scriptsgithub.com | AI Agent Workflow / Skill definition file (Markdown) / Open source on GitHub | Routes implementation to Codex (flat-rate, faster at code) while Claude retains architecture, design, and all verification |
| skills/opencode-fusion/SKILL.md at main · samirpatil2000/skillsgithub.com | AI Orchestration Skill / Skill definition document (Markdown) with bash script / Open source on GitHub | Three-phase pipeline: parallel model execution, structured judge analysis, and final synthesis—fully automated via a single bash script |
| GitHub - multica-ai/andrej-karpathy-skills: A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy’s observations on LLM coding pitfalls.github.com | AI Coding Guidelines / GitHub repository with markdown guideline file / open source, free on GitHub | Four principles target specific LLM coding failures: wrong assumptions, overcomplication, orthogonal edits, and lack of verifiable success criteria |
| Harness engineering: leveraging Codex in an agent-first worldopenai.com | Software Engineering Case Study / long-form article / free online | Generated one million lines of code over five months using only Codex with no human-written code. |
| Turning my laptop into a Search Relevance Judge with local LLMssoftwaredoug.com | Search Engineering Tutorial / long-form technical blog post with code repository / free online article with open source example repository | Local Qwen 2.5 LLM on a MacBook can evaluate hundreds of search result pairs per minute for free |
| Using LLM-as-a-Judge For Evaluation: A Complete Guide – Hamel's Blog - Hamel Husainhamel.dev | Technical Guide / long-form article / free online | Seven-step framework for building LLM-as-a-Judge evaluators |
| steipete/oracle: Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.github.com | Command‑line AI tool / open‑source command‑line interface / free open source on GitHub | Enables prompt submission with attached files to GPT‑5.1 or GPT‑5 Pro via API or browser mode |
A collection of bookmarks filtered by the tag "Prompt Engineering".
Showing 1-24 of 28 bookmarks tagged with "Prompt Engineering"
All Stories
<p>As a fun personal project, I have been working on a chatbot for answering general questions about my household on anything from maintenance questio...
