Skip to main content
affpapa.org
RU EN
skill CX Codex

plan-do-check-act — Claude Code skill: run a PDCA loop

Walks your agent through four PDCA phases: plan, do, check, act. Useful when you want to improve a process instead of guessing at it.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add NeoLabHQ/context-engineering-kit --skill plan-do-check-act -a claude-code
Open on GitHub

What plan-do-check-act does

The skill turns an improvement task into four phases. The agent states the problem and goal, captures baseline metrics, digs for root causes, and writes a hypothesis in the form "if we change X, Y improves". Then it designs the experiment and sets measurable success criteria.

In DO the change ships at small scale, with deviations and unexpected observations written down. CHECK measures against the criteria and the baseline, and says whether the hypothesis held. ACT decides what happens next: a win gets standardized through docs and checklists, a loss sends you back to a refined hypothesis and a fresh cycle.

Invoke it as /plan-do-check-act [improvement_goal]. Leave the goal out and the skill asks for it; the cycle number defaults to 1. For root-cause work the phases point at /why and /cause-and-effect — if those commands aren't in your setup, that analysis is on you.

What it helps with

  • find out why a build takes so long and record before/after numbers
  • write a testable hypothesis with success criteria instead of a vague goal
  • compare a change against the baseline rather than against a feeling
  • decide what to standardize after a win and what to roll back
  • start the next cycle when the last one only partly worked

Example prompts for the agent

  • /plan-do-check-act cut Docker build time
  • /plan-do-check-act reduce production bugs
  • run a PDCA cycle on code review speed, start from baseline metrics

How to install plan-do-check-act

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill plan-do-check-act -a claude-code

2Manually: put the files in place

~/.claude/skills/plan-do-check-act/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/plan-do-check-act/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about plan-do-check-act

How do I install plan-do-check-act in Claude?

Run npx skills add NeoLabHQ/context-engineering-kit --skill plan-do-check-act -a claude-code. The skill file lands at ~/.claude/skills/plan-do-check-act/SKILL.md.

What does plan-do-check-act do?

It drives the agent through the PDCA loop: define the problem and hypothesis, implement the change, measure the result, then standardize or adjust. You can repeat the cycle until the target is met.

Which commands does the skill rely on?

Its phases reference /why and /cause-and-effect for root-cause analysis. Without those commands, that part of the work stays manual.

Similar skills

All skills for Agent setup: memory, context, orchestration →
ce-plan EveryInc Builds structured technical implementation plans without touching production code. Useful when scoping features before passing work to execution agents. CX Codex +1 3,700installs speckit-plan github-spec-kit Executes implementation planning workflows inside spec-kit projects to produce research notes and interface contracts. CX Codex +1 2,500installs re0-plan LilMGenius Initializes a .re0 iteration folder and writes its starter files before re0-loop starts. Useful for structuring work and tracking quality gates in repositories using the paperthin skill suite. CX Codex +1 1,200installs plan owainlewis Turns an approved spec or brief into ordered vertical tasks for separate agent runs. It groups work into usable milestones with acceptance checks. CX Codex +1 747installs plan-writing davila7 Breaks work into short tasks, each with its own verification step. Useful for feature work, refactoring, or any multi-step job. 188installs make-plan thedotmack Turns a feature request into a phased plan built on real documentation instead of guesses. Useful right before you run do. CX Codex +1 7,200installs

From the same set64

analyze-issue Analyzes a GitHub issue, reviews related code, and saves a technical specification. Use it before starting work on a tracked task. CX Codex +1 1,200installs apply-anthropic-skill-best-practices A skill that walks you through Anthropic's official skill-authoring rules while you build your own. Useful once a skill file grows past a few screens. CX Codex +1 1,200installs create-agent Scaffolds autonomous subagents for Claude Code with valid YAML frontmatter, isolated contexts, and triggering examples. CX Codex +1 1,200installs do-competitively Runs three agents on the same task in parallel, scores their output with judges, then synthesizes the best result. Built for work where quality beats speed. CX Codex +1 1,200installs commit Automates Git commits using conventional commit messages and matching emojis. It checks the active branch, runs pre-commit linters, and suggests splitting complex diffs into atomic commits. CX Codex +1 1,200installs create-skill Helps create or update skills with TDD, using subagent scenarios to test an instruction before it is revised. CX Codex +1 1,200installs create-ideas This skill makes the model return six candidate answers with probabilities: three likely ones and three pulled from the tails of the distribution. CX Codex +1 1,200installs cause-and-effect Runs structured Ishikawa diagram investigations on development bottlenecks. Pinpoints root causes across technical and team factors. CX Codex +1 1,200installs

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