/crisp-teach
Teach CRISP your product once.
(What it does)
A one-time setup interview. /crisp-teach asks about five things (your product, users, design system, benchmarks, and known weaknesses) and saves the answers to .crisp.md in your project root.
Almost every other CRISP command reads that file automatically. Without it, you get generic design advice. With it, the advice is based on your users, your design system, and the products you want to be measured against.
(When to use it)
Run it once per project, before anything else. Run it again when:
- You are about to run your first
/crisp-auditor/crisp-reviewand.crisp.mddoes not exist yet - The product has changed direction: a new main user, a new job to be done, or a new stage
- You have adopted or replaced a design system and the documented tokens are out of date
- The products you compare yourself against have changed
(What you get)
- .crisp.md context file
- User personas and the jobs they need done
- Design system reference
- Products to match, and products to avoid copying
(Example output)
Illustrative example only. This is not a real audit.
(The interview)
Five sections, asked one at a time. Better context makes every later command more accurate, so the interview takes its time.
- Product: what it is, what stage it is at, and the main action users take
- Users: who they are, what they need the product to do, and what it costs them when it fails
- Design system: the library, the tokens, and any components that must not change
- Benchmarks: products to match, products to avoid copying, and the CRISP dimension that matters most right now
- Known weaknesses: UX problems you already know about, and anything audits should leave alone
(It's working if)
- Audit findings mention your documented users, not a generic persona
- Benchmark comparisons name the products you chose, not a default list
- Problems are weighted by the priority dimension you set
- You no longer re-explain the product at the start of every session
(Common questions)
Do I need to run /crisp-teach before the other commands?
No. Every command works without it, but the results are generic: default benchmarks, assumed users, and no knowledge of your design system. Each command reads project context first, so ten minutes of interview improves everything that follows.
Should I commit .crisp.md to the repository?
Yes. It is plain markdown and contains no secrets. Committing it means every agent session, for everyone on the team, starts from the same context. Treat it like any other project documentation.
How do I update the context later?
Run /crisp-teach again. It rewrites .crisp.md from a fresh interview. For a small correction, such as a changed benchmark or a new known issue, you can edit the file directly. It is written to be read and edited by people.
What happens if my answers are vague?
The output is only as sharp as your answers. If you cannot name a main user or a benchmark yet, say so. A documented gap is more useful to the other commands than an invented answer.
(Install)
Installs all fourteen CRISP skills and detects which AI coding tool you use. Run /crisp-teach once per project, before the other skills.
(Related skills)