# cobusgreyling/loop-engineering

Practical patterns, starters & CLI tools for loop engineering with AI coding agents. Design systems that prompt and orchestrate agents (inspired by Addy Osmani and Boris Cherny). Includes loop-audit, loop-init, loop-cost.

Repository: https://github.com/cobusgreyling/loop-engineering
Canonical: https://ross.abutalabs.com/products/loop-engineering
Homepage: https://cobusgreyling.github.io/loop-engineering/
Language: JavaScript
License: MIT
License Family: permissive
Topics: agentic-ai, ai-agents, claude-code, codex, devops-automation, github-actions, grok, llm, mcp, ai-coding, automation, loop-engineering, prompt-engineering, anthropic, claude, coding-agents, devtools
Last push: 2026-08-26T08:22:02+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 93, longevity 6
- inputs: {"age_days": 85, "days_push": 7, "days_rel": 44, "gap_med": 20, "n_releases_24m": 2}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10687, forks 1465 (observed 2026-08-28T04:10:43.747928+00:00)

## What it is
A collection of patterns, starter templates, and npm CLI tools for 'loop engineering' — designing automated loops that prompt and orchestrate AI coding agents like Claude Code, Codex, and Grok. It includes tools such as loop-audit, loop-init, and loop-cost that score repo readiness, scaffold agent loops, and track agent costs.

## Use cases
- audit a repo for readiness to run AI coding agent loops
- scaffold an automated agent loop in my project
- track and estimate the cost of AI coding agent runs
- set up scheduled agent runs with worktrees and verification
- learn patterns for orchestrating Claude Code and Codex without manual prompting
- add MCP connectors and sub-agent maker/checker splits to my workflow
- run AI coding agents from GitHub Actions on a schedule

## When to choose
- you use AI coding agents (Claude Code, Codex, Grok) and want structured, repeatable automation loops instead of ad-hoc prompting
- you want ready-made CLI tooling to audit, initialize, and monitor agent loops
- you want tool-agnostic patterns that work across multiple coding agents and GitHub Actions

## When to avoid
- you need a general-purpose agent framework for building custom LLM applications rather than coding-agent workflows
- you only prompt agents occasionally and don't need scheduled or automated loops
- you need a GUI or hosted platform rather than CLI tools and patterns

## Facets
- artifact type: cli-tool
- maturity: active
- function: cli, agent-framework, prompt-engineering, mcp, workflow-automation, developer-tools, ci-cd
- domain: developer-tools, large-language-models
- platform: cli, cross-platform
- tags: loop-engineering, claude-code, codex, agentic-coding, ai-coding-agents, npm-packages, prompt-orchestration, agent-workflows, worktrees, github-actions, ai-agents, automation, command-line, nodejs

## Member repositories
- cobusgreyling/loop-engineering (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:43.747928+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:18:10.691805+00:00, confidence not recorded.
  - readme: https://github.com/cobusgreyling/loop-engineering (fetched 2026-08-28T04:10:43.747928+00:00, sha aaa86d52a66d)
  - homepage: https://cobusgreyling.github.io/loop-engineering/ (fetched 2026-08-29T08:17:34.759388+00:00, sha e791e3a12a68)
- Data as of 2026-08-30T08:39:29.467469+00:00.