Skip to content

Contributing

Thanks for helping! This repo gets better with every real-world lesson people add. You don't need to code, and you don't need to have contributed to open source before.

How this repo works (in 30 seconds)

  • A skill is one Markdown file (SKILL.md) that teaches one growth task: goal, steps, a copy-paste AI prompt, and an example. Skills live in skills/<area>/<skill-name>/.
  • Areas group skills: strategy, research, case-study, listing, aso, launch, social, monetization, email. Each area has a README.md listing its skills.
  • Playbooks (playbooks/) chain several skills into a route, like a 6-week launch.
  • Everything is plain text, so you can edit it right on GitHub.

Pick how you want to help

Time What How
2 min Star the repo ⭐ at the top of the page
5 min Report what worked or didn't Open a Skill feedback issue
10 min Fix a typo, a broken link or an outdated rule Open the file on GitHub → ✏️ Edit → Propose changes (GitHub makes the fork and PR for you)
15 min Add a community, source or tool to a list Same ✏️ edit, e.g. the map in research-where-users-ask
30 min Share a real example with numbers Write a case study, edit a skill's Example output, or open an issue and we'll add it with credit
1–2 h Write a new skill Follow the guide below. Check the roadmap for skills nobody has claimed

Not sure? Ask in Discussions. No question is too basic.

Your first edit on GitHub (no tools needed)

  1. Open the file you want to change on github.com.
  2. Click the ✏️ pencil (Edit this file).
  3. Make your change. The Preview tab shows how it will look.
  4. Scroll down, write one line about what you changed, and click Propose changes, then Create pull request.
  5. We'll review it within about 48 hours, and may suggest small edits. That's normal, not a rejection.

Writing a new skill (step by step)

  1. Claim it. Open a Skill request issue (or comment on an existing one) so nobody writes the same skill twice.
  2. Copy the template. Create skills/<area>/<area>-<task>/SKILL.md from SKILL_TEMPLATE.md. The folder name must equal the name field, e.g. skills/aso/aso-competitor-audit/SKILL.md.
  3. Fill every section. Goal · When to use · Inputs · Steps · Prompt · Example output · Common mistakes · Related skills. Look at aso-keyword-research as a model.
  4. Test it. Run the prompt in an AI assistant on a real product. Does the output match your Example output?
  5. List it. Add the skill to the area README.md. If you can, also add it to .claude-plugin/marketplace.json, the router in skills/growth/SKILL.md and the root README.md. If you can't, maintainers will.
  6. Check it (optional, needs Python 3). Run python3 scripts/validate_skills.py. It checks names, sections and links. If you can't run it, just open the PR and we'll run it for you.
  7. Open a pull request and tick the checklist.

Tip: to see how your skill links to the others, open the skills/ folder in Markdown Everywhere (free, Mac, made by the maintainer) and look at the link map.

When is a skill "done"?

  • name matches the folder. description says what the skill does and when to use it ("Use when…").
  • Every section is there, with at least one copy-paste prompt fenced with four backticks (````).
  • The example is real, or starts with > **Illustrative example.**
  • Someone new to the topic can finish it in about an hour and ends up with something concrete.
  • Every step is an action ("Write 20 seed keywords"), not advice ("Think about keywords").
  • Facts like store rules, limits and prices link to an official source.

Style

  • English, plain words, short sentences. Write for someone smart but new to the topic.
  • Numbers beat adjectives: "30 characters", "2–4 weeks", "at least 12 testers".
  • Mentioning your own product is fine if it genuinely helps: say it's yours, once, at the end.

Using an AI to write or edit

Welcome. CLAUDE.md / AGENTS.md give the agent the rules. You're still responsible for every claim, so check the facts and test the prompt yourself.

The fine print

By contributing you agree that your contribution is licensed under the MIT License and that you follow the Code of Conduct. Contributors are credited in the skill's author field and its Credits section.