# Leonxlnx/unlazy

Anti-laziness skill for AI agents. Core: the Depth Tree method, which splits a task N layers deep and gives every leaf the full time budget of the whole task, so effort multiplies with depth. Grounded in 2025-2026 research on model laziness, underthinking and premature completion.

Repository: https://github.com/Leonxlnx/unlazy
Canonical: https://ross.abutalabs.com/products/unlazy
Language: JavaScript
License: MIT
License Family: permissive
Topics: ai-agents, claude, claude-code, llm, productivity, prompt-engineering, skill
Last push: 2026-08-24T16:46:40+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 1
- inputs: {"age_days": 24, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2545, forks 144 (observed 2026-08-28T04:06:57.252952+00:00)

## What it is
unlazy is a skill for AI coding agents (Claude Code, Codex CLI) that enforces completion discipline via the Depth Tree method, splitting tasks into layers with full time budgets per leaf. It includes a Node-based gate checker that runs and reverifies acceptance checks recorded in a GATES.md ledger.

## Use cases
- stop ai agents from finishing tasks prematurely
- force thorough multi-layer task decomposition for llm agents
- verify ai-generated code against runnable acceptance gates
- reduce underthinking and laziness in claude code workflows
- reverify completed work with automated checks
- install a discipline skill for ai coding assistants

## When to choose
- you use Claude Code or Codex CLI and agents cut corners on substantial tasks
- you want evidence-backed, re-runnable acceptance checks for agent work
- you need structured task decomposition with per-leaf effort budgets

## When to avoid
- you need a general-purpose agent framework rather than a discipline skill
- your workflow has no shell execution capability for gate checks
- you want a plug-and-play solution without reviewing shell commands in gate ledgers

## Facets
- artifact type: plugin
- maturity: active
- function: agent-framework, prompt-engineering, cli, testing, workflow-automation
- domain: large-language-models, developer-tools
- platform: cli, cross-platform
- tags: claude-code, claude-skills, ai-agent-discipline, depth-tree, task-decomposition, verification-gates, anti-laziness, skill, ai-agents, automation, nodejs

## Member repositories
- Leonxlnx/unlazy (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:57.252952+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 2026-08-30T02:24:36.630788+00:00, confidence not recorded.
  - readme: https://github.com/Leonxlnx/unlazy (fetched 2026-08-28T04:06:57.252952+00:00, sha ab2ddc6948e6)
- Data as of 2026-08-30T08:39:29.467469+00:00.