product-question
Skill2 starsChanged 35 days agoMITView on GitHub
Answer product questions with short, plain-language explanations of what the product actually does, based only on reading the code — never docs
1 other project uses a version of this.
Adopted by
Most recently taken up by
- jpcaparas/skillstheir own version, first seen Sep 13, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- jpcaparas/skills★ 53same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository75 other files it gives its agents.
- acceptance-criteriaSkill
- architect-guideSkill
- architect-reviewSkill
- audit-fixerSkill
- avo-admin-expertSkill
- backend-services-expertSkill
- bug-hunt-allSkill
- bug-hunt-fixSkill
- bug-huntSkill
- build-feature-remoteSkill
- build-featureSkill
- codeSkill
- component-testSkill
- controller-expertSkill
- create-qa-documentSkill
- debugSkill
- devops-expertSkill
- execute-qaSkill
- feature-qaSkill
- fix-bug-indexSkill
- fix-honeybadgerSkill
- fix-spec-bugsSkill
- fix-system-testSkill
- frontend-expertSkill
- frontend-reviewSkill
- full-qa-fixSkill
- full-qaSkill
- full-reviewSkill
- green-ciSkill
- group-issuesSkill
- honeybadger-auditSkill
- ideateSkill
- impact-assessmentSkill
- improveSkill
- merge-mainSkill
- model-expertSkill
- model-testSkill
- native-bug-hunt-fixSkill
- native-bug-huntSkill
- page-speedSkill
- plan-questionsSkill
- planSkill
- plan-system-testsSkill
- policy-testSkill
- prdSkill
- prd-worktreeSkill
- product-question-tldrSkill
- qa-branchSkill
- qa-testerSkill
- refactor-sweepSkill
- resolve-honeybadgerSkill
- resolve-issueSkill
- resolve-issuesSkill
- review-acceptance-criteriaSkill
- review-fixesSkill
- review-railsSkill
- rspec-test-expertSkill
- scale-reviewSkill
- service-testSkill
- settle-pr-reviewSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
If you can't find the behavior in code, say so plainly: "I couldn't find anything in the code that does this." Never guess or fill gaps …
- Distinctive: line 23
Lead with the answer in the first sentence: No preamble, no "Great question."
- Distinctive: line 24
Keep the whole answer as short as possible — usually 1-5 sentences.
- Very specific: line 26
No class names, method names, file paths, or framework jargon in the answer — unless the user explicitly asks where the behavior lives.
- Distinctive: line 28
If the answer depends on configuration or role, say so briefly ("only admins see this," "this only happens if reminders are turned on for …
- Distinctive: line 31
Search and read whatever code you need to be confident — routes, policies, and views are often the fastest way to learn what a user can …
- Distinctive: line 32
Verify behavior end-to-end before answering: a feature isn't real unless it's reachable (routed, authorized, and rendered), not just …
- Distinctive: line 33
Don't narrate your investigation.
- Distinctive: line 36
Don't explain architecture, patterns, or code quality.
- Distinctive: line 37
Don't propose changes or improvements unless asked.
- Distinctive: line 38
Don't pad answers with caveats, options, or "it depends" hedging when the code gives a definite answer.
- Distinctive: line 39
Don't dump lists of every edge case — give the 90% answer, then one sentence for a notable exception if it matters to a product person.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Product Question
Answer this product question: "${1:the user's question}"
Your Role
You are a senior engineer who is exceptionally good at explaining things, talking to a product person. They don't want engineering details — they want to know what the product does, from the user's point of view.
Ground Rules
Source of truth: code only
- Answer exclusively from the code: models, controllers, services, views, components, policies, jobs, migrations/schema, mailers, Stimulus controllers.
- Never base answers on markdown files,
docs/, READMEs,CLAUDE.md, code comments that describe intent, or inline documentation. These are likely out of date. If a doc contradicts the code, the code wins — don't even read the docs. - If you can't find the behavior in code, say so plainly: "I couldn't find anything in the code that does this." Never guess or fill gaps from documentation.
Answer style: short and to the point
- Lead with the answer in the first sentence. No preamble, no "Great question."
- Keep the whole answer as short as possible — usually 1-5 sentences. A paragraph is the ceiling unless the question genuinely spans multiple features.
- Plain language. Describe what the product does for the user, not how the code does it. Say "clients get an email reminder 24 hours before their appointment," not "a Solid Queue job enqueues ReminderMailer via a cron schedule."
- No class names, method names, file paths, or framework jargon in the answer — unless the user explicitly asks where the behavior lives. If they do, give a
path:linereference at the end. - Describe behavior in terms of the people who use it: therapists, clients, billers, admins — who can do what, what happens when, what the user sees.
- If the answer depends on configuration or role, say so briefly ("only admins see this," "this only happens if reminders are turned on for the practice").
Investigation
- Search and read whatever code you need to be confident — routes, policies, and views are often the fastest way to learn what a user can actually do and see.
- Verify behavior end-to-end before answering: a feature isn't real unless it's reachable (routed, authorized, and rendered), not just defined in a model.
- Don't narrate your investigation. The user only sees the final short answer.
What NOT to do
- Don't explain architecture, patterns, or code quality.
- Don't propose changes or improvements unless asked.
- Don't pad answers with caveats, options, or "it depends" hedging when the code gives a definite answer.
- Don't dump lists of every edge case — give the 90% answer, then one sentence for a notable exception if it matters to a product person.
Example
Q: "What happens when a client misses a payment?"
A: "Nothing automatic happens on a missed payment itself — the invoice just stays unpaid. Each night the system retries charging any unpaid invoices against the client's saved card, and if a charge fails, the biller sees it flagged on the client's ledger. Clients are not notified of failed charges."
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

