Loop Engineering is the practice of designing recurring systems for AI agents and coding agents. Instead of prompting an agent turn by turn, you build a loop that discovers work, delegates it to one or more agents, verifies the result against tests or other deterministic gates, persists state outside the model, decides what happens next, and runs again on a cadence, an event, or until a verifiable goal is reached. It sits above prompt, context, and harness engineering: those improve a single run, while loop engineering governs repeated agent work over time, including budgets, retries, escalation to humans, and stopping conditions.
agent wiki
Build the system that prompts your agents. A teaching repo for loop engineering: chapters, an annotated reading list, copy-paste prompts, a runnable example, and a portable agent skill
Build LLM systems you actually control. A free, open engineering book + course — from tokenization to serving your own models. Mechanisms, trade-offs, and numbers, not prompt tips
Provide a comprehensive collection of system design concepts and examples using `Mermaid diagrams` and `draw.io`.
A practical Loop Engineering playbook with workflows, safety gates, templates, and hand-drawn illustrations.
Loop Engineering für KI-Agenten — der loop-creator-Skill erkennt, ob sich ein Loop lohnt, baut ihn und prüft ihn (Loop-Doktor). Mit Landingpage zum Thema. MIT-Lizenz.
從會用 AI，進階到打造 AI 作業系統
A single-page (no build, no backend) knowledge base on Agentic AI: a 16-chapter Loop Engineering coursebook (EN/VI/ZH) plus a senior-level AI interview question bank (VI) — searchable, tag-filterable, deployable to any static host.
Loop Engineering-powered agent workflow for planning, generating, validating, repairing, and delivering editable PowerPoint decks.
A Band room that engineers a verification loop per task, then runs it — loop engineering done by a band of agents. Track 2, Band of Agents Hackathon.
An interactive, multi-page HTML book that teaches Agent Package Manager (APM)
A CI/CD-integrated cognitive control plane that gates pull requests behind interactive architectural quizzes to prevent developer deskilling and rubber-stamping of AI-generated code.
A self-describing agent kernel in which agents are programs that write programs.
一份面向中文读者的智能体环境工程开源手册，系统梳理 Prompt、Context、Harness、Loop 到 Agentic Environment Engineering 的演进，涵盖环境建模、合成、评测、协同演化与生产运行时治理。