The flagship. Understand the whole product, then surface prioritized opportunities across UI, UX, retention, engagement, community, and beauty.
Free & open · no signup · read-only — it ends by asking · nothing leaves your machine
You are working inside this repo. Mission: deeply understand this product, then surface a grounded, prioritized set of opportunities — new features, fixes, synergies, updates, iterations, improvements — across every surface: UI, UX, friendliness, helpfulness, community, retention, engagement, performance, beauty.
Discovery only, not implementation. Read anything, run read-only checks (tests, dev server, linters), but don't modify code. Your only write is the report file.
Phase 2 sweeps the codebase through every one of these, citing file and line for each finding.
hierarchy; spacing/type/color consistency; hover/focus states; motion; dark mode; polish
steps to core value; dead ends; confusing nav; missing undo/confirmations
empty states, sample data, guided first action, time-to-first-value
microcopy clarity, jargon, error messages that say what to do next, accessibility
smart defaults, suggestions, autofill, anticipating needs
loading/skeletons, optimistic updates, success confirmation, graceful failure
reasons to return: saved progress, history, streaks, digests, resurfacing past value
core-loop depth, personalization, progression, delight
sharing, collaboration, invites, comments, profiles, viral loops
existing features/data that combine into more than their parts; often the cheapest big wins
likely bugs, TODOs, dead code, perf drags, anything eroding trust
mobile responsiveness, offline, i18n, shareable links/SEO, PWA
One structured report at the repo root — or in reports/, if you keep one — the same shape every time, ready for a teammate — or the optional Studio — to act on.
Every Goal Prompt follows the same four steps, so results are consistent and repeatable — no matter which one you run.
Explore until you can answer from evidence, not guesses.
Find concrete evidence in code for each — never hypotheticals.
20–30 raw ideas → cut to the strongest 10–15.
Create IMPROVEMENTS.md at repo root.
Copy it and paste it into your agent inside the repo you want checked.
Install the goal plugin once — two commands — then just type /goal:product-improvement.
/plugin marketplace add GhostlyGawd/goal-prompts/plugin install goal@goal-promptsOr install only this Goal Prompt as /goal-product-improvement:
curl -fsSL https://goal-prompts.vercel.app/install | BRIEF=00 shLet an agent fetch it mid-conversation, or pull the raw Goal Prompt by URL.
https://goal-prompts.vercel.app/raw/00.mdNothing hidden — this is the whole Goal Prompt, verbatim. Read it in a minute, edit it, or copy it as-is.
# Goal: Product Improvement Discovery You are working inside this repo. Mission: deeply understand this product, then surface a grounded, prioritized set of opportunities — new features, fixes, synergies, updates, iterations, improvements — across every surface: UI, UX, friendliness, helpfulness, community, retention, engagement, performance, beauty. Discovery only, not implementation. Read anything, run read-only checks (tests, dev server, linters), but don't modify code. Your only write is the report file. ## Phase 1 — Understand the product Explore until you can answer from evidence, not guesses: - What is this product, who is it for, what core job does it do? - Current journey: entry → first value → habit → sharing/return? - What existing infrastructure, data, or half-built features are underused? Check: README/docs, package manifest, routes/screens, key components, API endpoints, schema, feature flags, analytics events, error handling, TODO/FIXME comments, recent git log. Getting the product wrong makes everything downstream wrong. ## Phase 2 — Audit through 12 lenses Find concrete evidence in code for each — never hypotheticals. 1. **UI & beauty** — hierarchy; spacing/type/color consistency; hover/focus states; motion; dark mode; polish 2. **UX & flows** — steps to core value; dead ends; confusing nav; missing undo/confirmations 3. **Onboarding** — empty states, sample data, guided first action, time-to-first-value 4. **Friendliness** — microcopy clarity, jargon, error messages that say what to do next, accessibility 5. **Helpfulness** — smart defaults, suggestions, autofill, anticipating needs 6. **Feedback & state** — loading/skeletons, optimistic updates, success confirmation, graceful failure 7. **Retention** — reasons to return: saved progress, history, streaks, digests, resurfacing past value 8. **Engagement** — core-loop depth, personalization, progression, delight 9. **Community** — sharing, collaboration, invites, comments, profiles, viral loops 10. **Synergies** — existing features/data that combine into more than their parts; often the cheapest big wins 11. **Fixes & felt debt** — likely bugs, TODOs, dead code, perf drags, anything eroding trust 12. **Reach** — mobile responsiveness, offline, i18n, shareable links/SEO, PWA ## Phase 3 — Generate wide, curate hard - 20–30 raw ideas → cut to the strongest 10–15 - Kill anything generic enough to fit any app - Every survivor cites repo evidence (paths, components, patterns) - Tag each: **FIX** (broken today) / **IMPROVE** (works, could be better) / **NEW** (doesn't exist) ## Phase 4 — Report Create `IMPROVEMENTS.md` at repo root: 1. **Product snapshot** — what, who, journey (5–8 sentences) 2. **Opportunity map** — impact × effort table 3. **Top 5 quick wins** (~a day each) + **Top 3 big bets** 4. **Full list** — each: Name · Tag+lens · Evidence · Proposal · Why it matters · Effort S/M/L · Impact L/M/H · Risks 5. **Sequence** — if only 3 ship first, which and why Start the report with today's date. If `IMPROVEMENTS.md` already exists from a previous run, read it first and lead with what changed since. ## Rules - No evidence, no idea - Prefer removing friction over adding features; synergies over net-new complexity - Think PM + designer + engineer at once: desirable, usable, feasible - If a `reports/` directory exists at the repo root, write the report there instead of the root. - Before asking, present the top findings as a ranked list in plain words - Report only — end by asking me which items to build
reports/ directory exists at the repo root, write the report there instead of the root.How user signal re-enters the product — corrections captured or evaporating, thumbs nobody reads, and the complaint-to-fix path.
Every message the product sends out of the app — transactional email, push, in-app — for whether it is timely, wanted, clear, and controllable.
The product's own search — whether users can find what they are looking for: relevance, recall, forgiveness, and what happens on zero results.
The whole arc of an account beyond signup — settings, export, plan changes, and deletion — and whether the product treats users with dignity, even leaving.
Runs inside these playbooks — curated sequences you can launch with one paste: