Absolute

Install

Clone https://github.com/maddhruv/absolute into a temporary folder, copy the absolute-init, absolute-work and absolute-spec folders into ~/.claude/skills/, then run /absolute-init at the project root.

This is third-party code. Review the repository files before installing.

What it does

The set has three groups. The one-time absolute-init interviews you, detects the stack and writes a config the other skills read. absolute-work drives the daily loop: interview, spec, a dependency-graphed task board and test-first execution in waves, with approval gates between phases; absolute-spec, absolute-ui, absolute-simplify and absolute-docs sit alongside. Five maintenance skills upgrade dependencies, scan for vulnerabilities, remove dead code, pay down lint and type debt and fix flaky tests. The skills never commit on their own.

Who it is for. For developers who want the agent to agree on design and plan before writing code.

Example request

/absolute-work Add Google and GitHub login, interview me about requirements first

Limitations

The skills are in English and target agents that support SKILL.md. The full absolute-work loop has many steps and confirmations, which raises token use. The skills run the project's tests and commands. The README install command points to repositories that do not exist, so installation is manual from the main repository.

How to disable. Delete the absolute-* folders from the agent's skills directory.

Security check

• Changes project code and upgrades dependencies
• Runs tests and build commands

README in short

The README explains which agent problems the set addresses and describes eleven skills by group: setup, daily loop and maintenance. It shows the absolute-work phases with gates and the .absolute-work/board.md state file. Installation uses npx skills add with flags for global install, agent selection and non-interactive mode. MIT licensed.

SKILL.md

---
name: absolute-work
version: 0.5.0
description: >
  End-to-end, phase-gated SDLC for AI coding agents: relentless design interview →
  reviewed spec → dependency-graphed task board → safe-wave TDD execution →
  verification → converge. Handles features, bugs, refactors, greenfield projects,
  planning breakdowns, and migrations.
  Triggers on "absolute work", "build this end-to-end", "plan and build",
  "break this into tasks", "grill me on this plan", "pick up this ticket",
  "run this migration".
category: workflow
tags:
  - workflow
  - sdlc
  - planning
  - tdd
  - spec
  - specification
platforms:
  - claude-code
  - gemini-cli
  - openai-codex
  - mcp
user-invocable: true
argument-hint: "[target]"
license: MIT
maintainers:
  - github: maddhruv
---

## Absolute Work: End-to-End AI Development Lifecycle

Absolute Work takes any unit of work — a ticket, a task, a plan, a migration — from
fuzzy intent to verified code. It is one continuous skill with **hard gates** between
phases: brainstorm a shared design, write and review a spec, decompose into a
dependency-graphed task board, then peel tasks off **one safe wave at a time** with
test-first verification. Nothing is assumed, nothing is silently expanded, and no code
is written until the design is approved.

The lifecycle has 6 phases:
**INTAKE & BRAINSTORM → SPEC → DECOMPOSE & PLAN → EXECUTE → VERIFY → CONVERGE**

Related