Skip to main content
affpapa.org
RU EN
skill Claude Code CX Codex Cursor GitHub Copilot

lean-build — Claude Code skill: what it does, how to install

Prevents overbuilding when adding new features or integrations. It sets strict acceptance bounds, cuts speculative abstractions, and stops immediately once proof passes.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add JuliusBrussee/caveman --skill lean-build -a claude-code
Open on GitHub

What lean-build does

The skill controls tasks with a high risk of overengineering. It derives observable acceptance criteria alongside explicit non-goals directly from the request. Then it traces entry points through invariant-owning layers. Forcing logic into a single file or relying on a fragile local patch is prohibited.

Speculative modes and extra configuration are stripped out unless acceptance strictly requires them. New dependencies or migrations require stating a material tradeoff. Once proof passes, execution stops. Turn this on when an agent overcomplicates simple product slices.

What it helps with

  • define observable acceptance criteria and explicit non-goals before coding
  • trace an end-to-end feature path across layers instead of patching a single file
  • eliminate speculative configurations, extra modes, and unneeded dependencies
  • stop agent execution immediately after running a focused proof

Example prompts for the agent

  • Implement the webhook slice using lean-build, omitting extra providers.
  • Add this API endpoint according to lean-build and stop once focused tests pass.

How to install lean-build

1Skills CLI (Vercel)

npx skills add JuliusBrussee/caveman --skill lean-build -a claude-code

2Manually: put the files in place

~/.claude/skills/lean-build/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/lean-build/

From the same set22

The whole set →
caveman JuliusBrussee Cuts filler and polite greetings from agent responses while keeping technical facts and code exact. Stays active until turned off. Claude Code CX Codex Cursor +1 591,500installs caveman-commit JuliusBrussee Generates terse Conventional Commits messages focused strictly on intent. Ideal for keeping git history clean without boilerplate or AI signatures. Claude Code CX Codex Cursor +1 414,700installs caveman-review JuliusBrussee Generates terse, single-line code review comments with exact line numbers, problems, and fixes. Eliminates conversational filler in pull request reviews. Claude Code CX Codex Cursor +1 410,700installs caveman-compress JuliusBrussee Rewrites CLAUDE.md and todo files into terse caveman-speak so the model reads fewer input tokens. The source gets overwritten; a readable backup stays out of tree. Claude Code CX Codex Cursor +1 410,300installs caveman-help JuliusBrussee A quick reference for Caveman modes and commands in Claude Code and other supported hosts. Use it when you need to control response style or find related skills. Claude Code CX Codex Cursor +1 401,800installs cavecrew JuliusBrussee cavecrew delegates code searches, small edits, and diff reviews to three subagents. Use it when keeping the main agent's context lean matters. Claude Code CX Codex Cursor +1 357,900installs caveman-stats JuliusBrussee Displays recorded output and cache-read tokens for Claude Code sessions or points to host native usage stats. Useful for tracking raw token consumption. Claude Code CX Codex Cursor +1 356,700installs caveman-explore JuliusBrussee A read-only skill locates relevant files and line ranges in a repository. Use it for cold-start orientation or broad searches. Claude Code CX Codex Cursor +1 106,500installs

Similar

ponytail-help DietrichGebert Shows a quick reference card for ponytail levels, skills, and configuration inside the chat. Useful when you need command triggers without leaving your editor. Claude Code CX Codex Cursor +1 42,300installs ponytail-debt DietrichGebert Turns every deliberate ponytail shortcut into one debt ledger: file, line, the ceiling the comment names, and the trigger for revisiting it. Claude Code CX Codex Cursor +1 42,200installs obsidian-cli kepano A skill for working with Obsidian vaults from the command line: read and create notes, search, edit properties, debug plugins and themes. Claude Code CX Codex Cursor +1 80,800installs find-skills vercel-labs Discovers and installs agent skills from skills.sh directly inside your coding assistant. Handy when you need domain-specific workflows for a project. Claude Code CX Codex Cursor +1 4,000,000installs grill-me Matt Pocock grill-me interviews you about a plan or design to expose gaps and sharpen decisions. Use it before implementation starts. Claude Code 1,424,476installs grill-with-docs Matt Pocock This skill interviews you about a plan or design and creates ADRs and a glossary along the way. Use it when decisions need sharper definition. Claude Code 1,219,756installs handoff Matt Pocock Creates a handoff document from the current conversation so another agent can continue the work. Use it when a task moves to a new session or agent. Claude Code 1,038,491installs grilling Matt Pocock Stress-test a plan, decision, or idea through structured rounds of questions. Use it when you need to expose assumptions before acting. Claude Code 979,543installs

Questions about lean-build

How do I install lean-build in Claude Code?

Run npx skills add JuliusBrussee/caveman --skill lean-build -a claude-code in your terminal. Alternatively, add the skill directly to ~/.claude/skills/lean-build/SKILL.md.

What is the purpose of lean-build?

It keeps feature development lean by setting explicit scope and non-goals. The agent avoids premature abstractions and halts as soon as acceptance checks pass.

How does lean-build handle architectural layers?

It requires an end-to-end path across responsible layers rather than forcing changes into a single file. Refactoring is required if a patch duplicates behavior or weakens ownership.

Source text

SKILL.md · Apache-2.0 
---
name: lean-build
description: Build feature work with high overbuilding risk. Use for new behavior, product slices, or integrations where repository reuse, strict scope, and an explicit stop condition matter.
---

# Lean build

Native Core's architecture-first simplicity remains mandatory. Turn feature into complete narrow outcome fitting system.

- Derive observable acceptance and explicit non-goals from request and repository.
- Trace entry point through layers owning invariants.
- Deliver coherent end-to-end path across responsible layers; never force work into one file, direct expression, or local patch.
- Reuse fitting seam. Refactor when patching duplicates behavior, weakens ownership, or hides root cause.
- Omit modes, providers, config, extensibility, and polish unless acceptance needs them.
- Add surface, dependency, service, config, or migration only for lifecycle design or acceptance; state material tradeoff.
- Keep work runnable; preserve Core safety.

Exercise path. Run focused proof. Stop when acceptance passes. Report only material omissions and trigger.

© JuliusBrussee · Apache-2.0 · JuliusBrussee/caveman

See also

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor