#187187th of 193 in Skills, plugins and prompts

loopy

Agent skill and catalog of bounded, self-checking prompt loops

Stars
3.2k
License
MIT
Last commit
Jul 2026
Language
JavaScript
Stars, last 19 weeks+3,198
9 Jun 2026: 0 stars16 Jun 2026: 0 stars23 Jun 2026: 1,485 stars30 Jun 2026: 2,192 stars7 Jul 2026: 2,519 stars14 Jul 2026: 2,664 stars21 Jul 2026: 2,796 stars28 Jul 2026: 2,887 stars4 Aug 2026: 2,945 stars11 Aug 2026: 2,986 stars18 Aug 2026: 3,019 stars25 Aug 2026: 3,045 stars1 Sept 2026: 3,071 stars8 Sept 2026: 3,106 stars15 Sept 2026: 3,121 stars22 Sept 2026: 3,140 stars29 Sept 2026: 3,148 stars6 Oct 2026: 3,160 stars13 Oct 2026: 3,172 stars20 Oct 2026: 3,198 stars

0 on 9 Jun 20263,198 on 20 Oct 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Loopy is an installable skill for Codex, Cursor and Claude Code that finds, audits, crafts, runs, debriefs, saves and publishes "loops": prompts with a verification check and a stopping rule. It works against the Loop Library website, a Cloudflare Worker-backed catalog of published loops that agents can also read as JSON and plain text. Saved loops go into a project-level LOOPS.md.

Who it is for: Developers using coding agents who want reusable, verifiable task loops

Strengths

  • Installs with one npx command for Codex, Cursor or Claude Code
  • Catalog is readable without the skill via catalog.json, catalog.txt and llms.txt
  • Run mode is bounded and returns a receipt with evidence and stopping reason
  • Saving and publishing require explicit approval of the exact preview

Weaknesses

  • Loop records are not in the repo; the catalog lives in a private Cloudflare database
  • Needs Node.js and npx for install
  • Recommendations and publishing depend on the hosted signals.forwardfuture.com site
  • No tagged releases

What it needs

  • no GPU
  • Needs Node.js, npx, Cloudflare Workers (website only)
  • Models: Codex, Cursor, Claude Code
  • port 4173

Also in Skills, plugins and prompts

See all 193
Also in Skills, plugins and prompts
RankProjectScore
1cavemanCuts token use in coding agents with terse replies and compressed inputs Live demo ↗ (opens in a new tab)114k starsApache-2.095 out of 100
2impeccableDesign skill with 24 commands and a detector for AI-generated frontends Live demo ↗ (opens in a new tab)84.1k starsApache-2.095 out of 100
3skillsAgent skills for planning, TDD, debugging and code review299.1k starsMIT94 out of 100
#4eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses285.1k starsMIT94 out of 100
#5oh-my-claudecodeMulti-agent orchestration plugin and CLI for Claude Code40.1k starsMIT93 out of 100