Fable 5 Is Gone. The System Prompt Isn't.
Claude Fable 5 lasted about 72 hours. The leaked system prompt is still public, and you can load it onto another model.
阅读文章文章
来自构建循环的笔记。
面向使用 Codex、Claude Code、Cursor 和各类 Agent 的开发者，分享实测过的 AI 编程工作流、编程 Agent 现场笔记和真实的大模型评测。
How to make AI-assisted writing sound less like slop
Humanizer is an open-source agent skill that catches 33 AI writing patterns. Pair it with your own voice samples and ship cleaner drafts.
阅读文章Fable 5 One-Shotted This 3D Globe Visualization
One screenshot and one prompt. Fable 5 rebuilt a production-looking Three.js globe dashboard in a single pass.
阅读文章Claude Fable 5 Is the Best Model You Can Use Right Now. Save It for the Hard Stuff.
Claude Fable 5 leads on hard coding benchmarks, but the cost is high. Send the hard work to it and keep everyday tasks on cheaper models.
阅读文章Loop Engineering: A 14-Step Roadmap from Prompter to Loop Designer
A practical roadmap for deciding whether to build an agent loop, then adding automation, state, verification, tools, and security controls in the right order.
阅读文章Loop Engineering: The Next Layer After Prompt Engineering
Peter Steinberger and Boris Cherny both say the job is designing loops, not writing prompts. Feedback, stop conditions, and token cost decide if it works.
阅读文章A Practical Guide to Using /goal in Codex
Use Codex goals as completion contracts with explicit outcomes, verification, constraints, iteration rules, and safe stop conditions.
阅读文章Codex Is No Longer Just a Coding Tool
OpenAI repositioned Codex for knowledge work with role-specific plugins, hosted Sites, and Annotations. Non-developers are growing fastest.
阅读文章MiniMax M3: Opus-Level Coding, DeepSeek Pricing
MiniMax M3 launched with 1M context, near-Opus coding scores, DeepSeek-like API pricing, and two ways to try it immediately.
阅读文章Used Codex to figure out what was eating my Mac's storage
A read-only Codex prompt that turned macOS System Data into a prioritized storage map, with 180 GB reclaimable on a 1TB Mac.
阅读文章