Learn to prompt through your everyday work.

Same AI. Clearer direction.

Same AI. The second prompt names the audience, the deliverable, the evidence and the quality bar.

Learn from the work you just did.

When you finish a task, the coach can offer one small lesson from it. You can stop after any step.

• Practice

  Write your own change and predict its effect, then see what changed.

• Example

  See a brief demonstration on your task.

• Lesson

  Learn the principle using your task, then practise if you want.

• Later

  Continue your work, with no penalty or backlog.

No improvement is a valid result. A short prompt can be enough.

See where your past prompts cost you turns.

On request, it looks back at your recent work for the follow-up that says “no, I meant…”, an interrupted run, a prompt you edited, or a rule you had to repeat.

• A casebook of your own prompts, each with one edit that would have helped, and lesson suggestions.
• It asks twice: before reading, and again before scoring.
• Your history stays with its provider. Claude reviews Claude exports; ChatGPT and Codex review ChatGPT exports.
• In Claude Code or Codex, results stay in a local folder you choose and expire automatically.

Carry your learning between chats.

A Learning Card, LEARNING.md, records what you practised and what you used on your own.

• You keep it. Save it and attach it in your next chat, in ChatGPT or Claude.
• Nothing syncs on its own. Without the current card, the coach can't know your learning history.
• Help and independent use stay apart. Reading an example doesn't count as learning the skill.

The optional course: how a lesson works

For a structured path: an assessment, then thirty guided lessons, each built on a task from your own working life. Everyday lessons don't advance the course.

• Write

  Your prompt for a real task.

• Run

  It runs exactly as written; it is not quietly improved first.

• Compare

  Against one targeted change, on the same material.

• Improve

  Name the change that mattered and get a score.

Thirty lessons, four stages

• Diagnose

  A short assessment records your starting level and the tasks you want help with.

• Control

  Eleven language levers: the ordinary parts of language that control what gets made, in what scope, and to what standard.

• Build

  Examples, output formats, multi-step tasks, self-review, evaluation criteria and reusable templates.

• Harden

  A capstone prompt for one of your real tasks, tested on a case it wasn't designed for.

Who it's for

Anyone who already asks AI for help. No coding knowledge is required.

• A teacher turns source material into a lesson plan with a defined age level and objective.
• A lawyer extracts obligations, dates and exceptions, keeping conclusions tied to the text.
• A manager turns meeting notes into decisions, owners and deadlines.
• A researcher separates evidence from inference.

You bring the professional judgment.

Install one plugin to start.

What you need

• Claude, ChatGPT, Claude Code or Codex. Prompting Wizard is one plugin for all four. Plugin uploads in the Claude app need a paid plan; in ChatGPT, a local plugin is shared through a workspace admin.
• A task from your own work. The coach and the lessons use it, so keep one in mind.
• A folder to practise in. See the note below the prompt.

Download for your app

• Claude (web, desktop, Cowork, Claude Code): prompting-wizard-claude.zip. Upload it in Customize → Plugins; it also appears in Claude Code when you sign in with the same login.
• ChatGPT and Codex: prompting-wizard-openai.zip. Register it as a local plugin in the ChatGPT desktop app, then a workspace admin can publish it; in Codex you can also add the marketplace below.

Then say “Use Prompting Wizard to teach me through my work; no interview,” or “Start the Prompting Wizard course.” Everyday coaching has passed the device walkthrough in Claude on the web. See plugin setup.

Or paste one prompt into Claude Code or Codex

It loads Prompting Wizard and offers everyday learning, the structured course, or later.

Install or load Prompting Wizard from:

https://github.com/Entuvo/prompting-wizard/tree/main/prompting-wizard

This setup is for Claude Code or Codex. First determine which of those hosts you are running in and which capabilities it actually provides. Use its native skill, plugin, marketplace, repository, or filesystem installation method when available.

If persistent installation is unavailable but you can read GitHub, load Prompting Wizard for this chat by reading AGENTS.md and then SKILL.md from that repository path. Follow those instructions directly.

Do not overwrite an existing installation without asking me. Never claim the skill is installed unless the host confirms it. Do not stop after explaining installation or creating a package. Once Prompting Wizard is available, offer everyday learning without an interview, the structured course, or later. Honor my choice.

For everyday learning, accept my current LEARNING.md if supplied. After I accept a learning record, return the complete updated card for me to retain unless a durable file write succeeds. Never claim it is saved or synced without that write.

For the structured course, if this environment cannot preserve files between chats, accept an attached PROGRESS.md and return the complete updated PROGRESS.md after every state change.

If you can neither install the skill nor read its files from GitHub, state exactly which capability is missing. Do not fabricate a successful installation.

Using commands instead?

/plugin marketplace add Entuvo/prompting-wizard
/plugin install prompting-wizard@entuvo-prompting

codex plugin marketplace add Entuvo/prompting-wizard

npx skills add Entuvo/prompting-wizard