ArchestackBest of Vibe Coding: List on GitHub ↗ (opens in a new tab)

#144144th of 193 in Skills, plugins and prompts

unlazy

Agent skill that requires runnable acceptance gates before work is reported complete

Repository on GitHub ↗ (opens in a new tab)
Stars
4k
License
MIT
Last commit
Sep 2026
Language
JavaScript
Stars, last 10 weeks+4,011
11 Aug 2026: 24 stars18 Aug 2026: 278 stars25 Aug 2026: 2,400 stars1 Sept 2026: 2,919 stars8 Sept 2026: 3,145 stars15 Sept 2026: 3,358 stars22 Sept 2026: 3,487 stars29 Sept 2026: 3,739 stars6 Oct 2026: 3,843 stars13 Oct 2026: 3,921 stars20 Oct 2026: 4,035 stars

24 on 11 Aug 20264,035 on 20 Oct 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

unlazy is a skill for Claude Code and Codex CLI that has the agent write an acceptance ledger (GATES.md) of CHECK/EXPECT shell commands, then runs them with a Node checker and reports only what passes. It also covers scoped multi-leaf orchestration with ownership leases, and an optional Claude Code Stop hook that blocks exit while gates are unmet. Requires Node 16+, no third-party runtime packages.

Who it is for: Developers running Claude Code or Codex agents on long, multi-step tasks

Strengths

  • Gates pass only on exit 0 plus EXPECT match, with definition-bound evidence
  • Checks need explicit per-command approval, stored outside the repository
  • No third-party runtime dependencies; Node 16 or newer, with a test suite
  • Installs through the skills CLI for several agents, plus manual Claude Code and Codex paths

Weaknesses

  • Proves only the command oracle you write; weak gates give false assurance
  • Approval is consent, not a sandbox; checks run with ambient filesystem and network access
  • Version 2.1.0 is unreleased; README advises pinning a commit
  • Earlier benchmark claims lack reproducible artifacts; the README calls them historical only

What it needs

  • no GPU
  • Needs Node.js 16+
  • Models: Claude Code, Codex CLI

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