Improve an App
Product, UX, copy, and in-app persuasion — fix what users feel
Takes you from a shipped app that works but feels rough to a product that fits the job, flows without friction, reads clearly, and persuades honestly. It orchestrates its nine skills phase by phase, asking you the decision questions at every step and recording every result in your project's docs/ folder so the journey resumes across sessions.
What is the Improve an App journey?
Some apps work and still disappoint. Nine phases close that gap: re-anchor on the job users hire the app for, strip usability friction, fix hierarchy and microinteraction detail, rewrite in-app copy, make upgrade moments persuade honestly, and face a final review — each answer stored in docs/ for the next session.
One prompt runs it
npx skills add wondelai/skills/improve-app --global This skill is compatible with Claude, Claude Code, Claude Cowork, Codex, Cursor, OpenClaw, Hermes Agent, and other agentskills.io-compatible agents.
Example prompts for the Improve an App journey
The app works but feels rough and users say it's confusing — improve the experience end to end. using improve-app skill
Experience overhaulRun the improve-app journey; start by checking what job users hire us for. using improve-app skill
Journey kickoffContinue improving the app from where the tracker says we stopped. using improve-app skill
Resume journeyWhen to use Improve an App — and when not to
Reach for it when
- Re-anchor the app on the job users actually hire it for
- Remove usability friction with severity-ranked audits before any visual polish
- Fix hierarchy, spacing, and microinteraction detail so screens feel professional
- Rewrite in-app copy (onboarding, empty states, errors, CTAs) so it lands
- Make paywalls and upgrade prompts persuade honestly, then face a cold final review
Reach for something else when
- The code is the problem, not the experience — Improve Code Quality works on tests, structure and reliability.
- Users are happy but too few — Grow an App attacks retention and activation rather than polish.
- You want one severity-ranked usability audit and nothing more — UX Heuristics does that on its own.
- There is no app yet to improve — Create an App sequences validation and architecture from the start.
Frequently asked questions
How do I install the Improve an App journey?
Run npx skills add wondelai/skills/improve-app --global. It takes about 30 seconds and needs no account. The Improve an App skill then works in Claude, Claude Code, Claude Cowork, Codex, Cursor, OpenClaw and Hermes Agent — anything that reads the open agentskills.io format — and your agent loads it on its own when a task calls for it. It is free and MIT-licensed, and the source is at https://github.com/wondelai/skills.
Which skills does the Improve an App journey run, and what does it produce?
It orchestrates 9 skills across 9 phases — Jobs to Be Done, UX Heuristics, Design of Everyday Things, Refactoring UI, Microinteractions, Made to Stick, Influence Psychology, High Perf Browser, Steve Jobs Design Review — asking you one decision question per phase. Each answer is written to your project's docs/ folder (docs/CUSTOMER.md, docs/DESIGN.md, docs/EXPERIMENTS.md, docs/POSITIONING.md, docs/PRODUCT.md, docs/IMPROVE-APP-PLAN.md), so the journey survives across sessions and you keep the reasoning, not just the output.
Why start with Jobs to Be Done rather than a UX audit?
Because polish applied to the wrong job produces a beautiful app that still underdelivers. The first phase is a gate: it establishes what users hire the app for and where it falls short, into docs/CUSTOMER.md. If the answer is that the app serves a job nobody has, no amount of microinteraction work will rescue it — and you will have learned that in a day rather than a quarter.
Where does this journey stop and Improve Code Quality start?
At the surface. This one works on everything a user can perceive — flows, affordances, hierarchy, motion, wording, paywall honesty, perceived speed. Improve Code Quality works on what they cannot: tests, structure, dependency boundaries, production resilience. They share no phases and no docs/ files, so running them in either order is fine. Start here if churn and confusion are the symptoms.
The persuasion phase makes me uneasy. Is this dark patterns?
The decision question is whether the paywall and upgrade moments persuade honestly, and the Influence Psychology phase is explicit about the line: reciprocity, social proof and scarcity applied to things that are actually true. Manufactured countdowns and fake stock counts fail the phase. The closing Steve Jobs review is a second check — an experience that tricks people does not pass as insanely great.
Can I run only the visual phases?
Yes. Name the ones you want and the journey runs UX Heuristics, Design of Everyday Things, Refactoring UI and Microinteractions as a block, writing findings to docs/DESIGN.md and docs/EXPERIMENTS.md. You give up the gate that checks you are polishing the right thing, and the copy and persuasion phases that often explain why a screen underperforms. The tracker keeps the skipped phases available.
Install Improve an App
Free, open-source, and ready in 30 seconds.
npx skills add wondelai/skills/improve-app --global MIT Licensed · Works with Claude, Claude Code, Claude Cowork, Codex, Cursor, OpenClaw, Hermes Agent & other agentskills.io agents · No account needed
Don’t guess your AI engineering level.
Measure it.
AI Developer Scorecard
How production-grade is your AI engineering?
Twenty-five questions across the practices that separate vibe-coding from production-grade engineering. Instant score, per-section breakdown, and a 30/60/90-day playbook.
Score 0–75 · 30/60/90-day playbook Score yourself For CTOs & foundersCTO Scorecard
Is your engineering team ready for AI at scale?
Twenty-five questions on how your org adopts AI — adoption, governance, automation, and ROI. See where you sit on the path from Reactive to Strategic Leader.
Score 0–75 · Reactive → Strategic Leader Score your orgWe build the skills you already use. Now we’ll build yours.
Sprints from $3K · shipped to production, or you don’t pay the final milestone.