AI Agents
pi
A TypeScript toolkit for building agents: unified LLM API, loop, TUI, and CLI.
★ 113.3k starsMIT
View guide →Loop Engineering — practical patterns, starters and a CLI for designing the loops that prompt and orchestrate AI coding agents, with a Loop Ready score
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.
Browse the pattern library — daily triage, PR babysitter, thin loop and more — and choose by cadence and cost.
Run the loop CLI init with your pattern and tool to generate the starter files in your repo.
Run doctor and the audit to get your Loop Ready score, then iterate on the loop instead of the prompts.
Each pattern documents its cadence, Week 1 deliverable and cost level, so loops are designed deliberately.
One front door — @cobusgreyling/loop — with init, doctor, status, audit and cost; the older single-purpose packages stay supported.
The audit scores how ready a loop is, turning agent operations into something measurable.
The cost command reports spend per pattern and autonomy level before a loop runs away with the budget.
Initialise the daily-triage pattern for Claude Code in the current project.
npx @cobusgreyling/loop init . --pattern daily-triage --tool claudeDoctor verifies the loop is wired correctly.
npx @cobusgreyling/loop doctor .Report spend for the pattern at a given autonomy level.
npx @cobusgreyling/loop cost --pattern daily-triage --level L1Cobus Greyling; the showcase site is cobusgreyling.github.io/loop-engineering.
Yes, MIT-licensed; the CLI is published on npm. Agent usage itself runs on your own model subscriptions or keys.
No key for the patterns or CLI; the agents your loops drive use your existing tooling and accounts.
Prompt libraries optimise single prompts. Loop Engineering optimises the recurring system around the agent — cadence, handoffs, audit and cost — and scores it.
Run npx @cobusgreyling/loop init . --pattern daily-triage --tool claude in a project, then npx @cobusgreyling/loop doctor . to verify it.
A TypeScript toolkit for building agents: unified LLM API, loop, TUI, and CLI.
Mobile app automation and verification built for AI coding agents — CLI, MCP server, and Node API
Babysitter enforces the process: deterministic, quality-gated orchestration for fleets of coding agents.
Compression gateway that cuts AI coding agents' token bills up to 85%.
Test your agents against prompt-injection attacks.
A CLI that writes and maintains agent-ready docs for your codebase.
The most-starred other AI Agents projects in the LibreAI catalogue — the realistic shortlist if Loop Engineering is not the right fit.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Free and open — grab the source, star the repo, and start creating.