AI Agents
pi
A TypeScript toolkit for building agents: unified LLM API, loop, TUI, and CLI.
★ 113.3k starsMIT
View guide →// agents · cobusgreyling/loop-engineering
Loop Engineering
Loop Engineering is an open-source ai agents project by cobusgreyling (cobusgreyling/loop-engineering on GitHub). It has 11,554 GitHub stars and 1,572 forks, is released under the MIT license, and was last updated on GitHub on 15 October 2026.
Loop Engineering — practical patterns, starters and a CLI for designing the loops that prompt and orchestrate AI coding agents, with a Loop Ready score
★ 11.6k stars⑂ 1.6k forksMITupdated 2026-10
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
so here's a placeholder. The guide below is complete regardless.
// project facts
Project facts
- Repository
- cobusgreyling/loop-engineering
- GitHub stars
- 11,554
- Forks
- 1,572
- License
- MIT
- Last updated on GitHub
- 15 October 2026
- Category
- AI Agents
- Listing
- Full install guide
- Official site
- Official site
// how it works
How it works
- Pick a pattern
Browse the pattern library — daily triage, PR babysitter, thin loop and more — and choose by cadence and cost.
- Scaffold it with the CLI
Run the loop CLI init with your pattern and tool to generate the starter files in your repo.
- Audit and improve
Run doctor and the audit to get your Loop Ready score, then iterate on the loop instead of the prompts.
// features
Why people love it
Patterns, not prompts
Each pattern documents its cadence, Week 1 deliverable and cost level, so loops are designed deliberately.
A unified CLI
One front door — @cobusgreyling/loop — with init, doctor, status, audit and cost; the older single-purpose packages stay supported.
Loop Ready score
The audit scores how ready a loop is, turning agent operations into something measurable.
Cost visibility built in
The cost command reports spend per pattern and autonomy level before a loop runs away with the budget.
Advertisement
// run it
Run it yourself
- Scaffold a loop
Initialise the daily-triage pattern for Claude Code in the current project.
npx @cobusgreyling/loop init . --pattern daily-triage --tool claude - Check the setup
Doctor verifies the loop is wired correctly.
npx @cobusgreyling/loop doctor . - Watch the cost
Report spend for the pattern at a given autonomy level.
npx @cobusgreyling/loop cost --pattern daily-triage --level L1
// use cases
What you can build
- Running a daily triage loop over issues and CI instead of manual prompting
- Letting an agent land and babysit PRs under a defined, scored pattern
- Standardising how a team operates coding agents across repos
- Auditing an existing agent setup and raising its Loop Ready score
Advertisement
// faq
Questions, answered
Who maintains Loop Engineering?
Cobus Greyling; the showcase site is cobusgreyling.github.io/loop-engineering.
Is it free?
Yes, MIT-licensed; the CLI is published on npm. Agent usage itself runs on your own model subscriptions or keys.
Does it need a paid API key?
No key for the patterns or CLI; the agents your loops drive use your existing tooling and accounts.
How does it compare to a well-known alternative?
Prompt libraries optimise single prompts. Loop Engineering optimises the recurring system around the agent — cadence, handoffs, audit and cost — and scores it.
How do I get started?
Run npx @cobusgreyling/loop init . --pattern daily-triage --tool claude in a project, then npx @cobusgreyling/loop doctor . to verify it.
// keep exploring
Similar tools
AI Agents
agent-device
Mobile app automation and verification built for AI coding agents — CLI, MCP server, and Node API
★ 4.9k starsMIT
View guide →AI Agents
babysitter
Babysitter enforces the process: deterministic, quality-gated orchestration for fleets of coding agents.
★ 1.8k starsMIT
View guide →AI Agents
paritok-4b-v1
Compression gateway that cuts AI coding agents' token bills up to 85%.
★ 1.5k starsApache-2.0
View guide →AI Agents
agentdojo
Test your agents against prompt-injection attacks.
★ 897 starsMIT
View guide →AI Agents
openwiki
A CLI that writes and maintains agent-ready docs for your codebase.
★ 17.0k starsMIT
View guide →// alternatives
Alternatives in AI Agents
The most-starred other AI Agents projects in the LibreAI catalogue — the realistic shortlist if Loop Engineering is not the right fit.
- openclaw★ 391.6k
A personal AI that really does things — on any OS, with your data staying yours.
- superpowers★ 294.9k
An agentic skills framework & software development methodology that works.
- ECC★ 275.2k
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
- hermes-agent★ 252.0k
The agent that grows with you — a Python framework for adaptive AI agents.
- deepseek-harness★ 245.5k
DeepSeek's plugin-based agent harness where everything is a plugin.
// more in agents
More in AI Agents
Nearby AI Agents projects in the catalogue, closest in stars to Loop Engineering — keep exploring the shelf either side of it.
- pest★ 11.7k
The elegant testing framework for PHP developers and AI agents.
- holaOS★ 11.5k
Local-first agentic workspace where Claude Code, Codex and a built-in agent share memory, apps and tools.
- phoenix★ 11.7k
Observability and evaluation for AI apps — see what your agents do.
- ten-framework★ 11.1k
Open-source framework for conversational voice AI agents
- MemOS★ 11.8k
Self-evolving memory OS for LLMs and agents: persistent graph memory, hybrid retrieval, skill reuse.
- spring-ai-alibaba★ 11.0k
Agentic AI Framework for Java Developers
Get Loop Engineering
Free and open — grab the source, star the repo, and start creating.