start
Skill40 starsChanged 7 months agoMITView on GitHub
First-time onboarding for the mobile app studio template. Asks where the user is starting from, classifies them, and routes them to the right next skill. Use this on the very first session, or whenever the project feels unmoored.
130 other projects use a version of this.
Adopted by
Most recently taken up by
- ansvisor/ansvisortheir own version, first seen Nov 8, 2026
- dsifry/metaswarmtheir own version, first seen Nov 8, 2026
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tooltheir own version, first seen Nov 8, 2026
- paulocsb/worldcup-fifa-pooltheir own version, first seen Nov 8, 2026
- srinator22/New-Projecttheir own version, first seen Nov 8, 2026
- Memolok/Claude-plugintheir own version, first seen Nov 6, 2026
Used with
Projects with it also use setup-engine, team-audio and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
CoralGameSkillsetup-engine
Configure the project's game engine and version. Pins the engine in CLAUDE.md, detects knowledge gaps, and populates engine reference docs via WebSearch when the version is beyond the LLM's training data.
Also in 13 projects that use this
CoralGameSkillteam-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
Also in 13 projects that use this
- +3 more from CoralGame
Other versions130 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 130
- anthropics/knowledge-work-plugins★ 35ksame name
- Donchitos/Claude-Code-Game-Studios★ 27ksame name
- openai/plugins★ 7.8ksame name
- SterlingChin/marvin-template★ 1ksame name
- educlopez/ui-craft★ 441same name
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tool★ 432same name
- dsifry/metaswarm★ 426same name
- pixel-cellar/Claude-Code-Game-Studios★ 324same name
- ansvisor/ansvisor★ 233same name
- alexeykrol/claude-code-starter★ 194same name
120 more not listed
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 repository77 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- design/CLAUDE.mdCLAUDE.md
- docs/CLAUDE.mdCLAUDE.md
- src/CLAUDE.mdCLAUDE.md
- tests/CLAUDE.mdCLAUDE.md
- adoptSkill
- architecture-decisionSkill
- architecture-reviewSkill
- asset-auditSkill
- asset-specSkill
- balance-checkSkill
- brainstormSkill
- bug-reportSkill
- bug-triageSkill
- changelogSkill
- code-reviewSkill
- consistency-checkSkill
- content-auditSkill
- create-architectureSkill
- create-control-manifestSkill
- create-epicsSkill
- create-storiesSkill
- day-one-patchSkill
- design-bibleSkill
- design-reviewSkill
- design-systemSkill
- dev-storySkill
- estimateSkill
- extractSkill
- gate-checkSkill
- helpSkill
- hotfixSkill
- launch-checklistSkill
- localizeSkill
- map-systemsSkill
- milestone-reviewSkill
- onboardSkill
- patch-notesSkill
- perf-profileSkill
- prd-reviewSkill
- project-stage-detectSkill
- propagate-design-changeSkill
- prototypeSkill
- qa-planSkill
- quick-designSkill
- regression-suiteSkill
- release-checklistSkill
- retrospectiveSkill
- reverse-documentSkill
- review-all-prdsSkill
- scope-checkSkill
- security-auditSkill
- setup-frameworkSkill
- skill-improveSkill
- skill-testSkill
- smoke-checkSkill
- soak-testSkill
- sprint-planSkill
- sprint-statusSkill
- story-doneSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
16 rules. 0 are common, 6 are distinctive, and 10 are very specific to this project.
0
Common
Most of the 117 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
10
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Do not run when the project is mid-sprint and the user has a clear active task — /help or /sprint-status are better.
- Distinctive: line 44
Do this without narrating.
- Very specific: line 47
Framework configured?: Read .claude/docs/technical-preferences.md.
- Very specific: line 60
Use AskUserQuestion once. Do not preload assumptions.
- Distinctive: line 69
Wait for their pick. Do not proceed otherwise.
- Distinctive: line 101
Ask them to share the theme in a sentence or two (free text).
- Very specific: line 108
Ask them to give the one-liner (free text — open response, not AskUserQuestion).
- Very specific: line 138
Check if production/review-mode.txt already exists.
- Very specific: line 140
If it exists: Read and report — "Review mode is set to [value]." Move to Phase 4.
- Very specific: line 142
If it does not exist: Use AskUserQuestion:
- Very specific: line 149
Write the choice to production/review-mode.txt immediately:
- Very specific: line 154
Create the production/ directory first if it is missing.
- Very specific: line 160
Use AskUserQuestion to confirm the recommended first step.
- Distinctive: line 167
If the user picks the alternative, accept their input and adjust.
- Distinctive: line 208
Ask first; never assume the user's state.
- Distinctive: line 211
If the user's situation does not fit a tab, listen and adapt.
“Common” is measured: kept by at least 6 in 10 of the 117 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Guided Onboarding (Mobile App Studio)
Single file written by this skill: production/review-mode.txt (only when the user picks a review mode in Phase 3b).
This is the entry point. It assumes nothing — not the framework, not the idea, not the user's experience level. It looks at what is on disk, asks the user where they are, then hands off.
Purpose / When to Run
Run when:
- The repo has just been cloned and nothing has been authored yet
- The user types
/startto re-orient - A returning collaborator wants a clean route into the right next step
Do not run when the project is mid-sprint and the user has a clear active task — /help or /sprint-status are better.
Inputs
.claude/docs/technical-preferences.md— to detect whether a framework has been pinneddesign/concept.md— concept doc presencedesign/prd/— PRD countproduction/review-mode.txt— existing review-mode settingproduction/sprints/andproduction/epics/— production-stage signalsrc/,ios/,android/,lib/— code presence
Outputs
production/review-mode.txt— only written if not present and the user makes a choice in Phase 3b
Phase 1: Silent State Probe
Do this without narrating. The findings only inform recommendations; they are not the conversation opener.
Check:
- Framework configured? Read
.claude/docs/technical-preferences.md. If the Framework field still contains[TO BE CONFIGURED], the framework is not pinned. - Concept exists? Glob for
design/concept.md. - PRDs exist? Count files matching
design/prd/*.md. - Source code present? Glob for any of:
src/**/*.ts,src/**/*.tsx,lib/**/*.dart,ios/**/*.swift,app/src/main/**/*.kt,**/*.swiftexcluding Pods. - Production work in flight? Count files in
production/sprints/andproduction/epics/. - Review mode set? Glob
production/review-mode.txt.
Hold these as facts to validate the user's self-assessment in Phase 4.
Phase 2: Ask Where the User Is
Use AskUserQuestion once. Do not preload assumptions.
- Prompt: "Welcome to the mobile-app studio template. Before recommending anything, where are you with the app right now?"
- Options:
A) No idea yet— I want to figure out what to build.B) Loose theme— I have a fuzzy sense (e.g., "a tracker for runners", "a calmer Instagram") but no spec.C) Clear pitch— I can give you a one-liner: target user, core job, rough mechanic.D) Existing app— There is already code, screens, or planning. I want to organize it.
Wait for their pick. Do not proceed otherwise.
Phase 3: Route Based on Answer
If A: No idea yet
- Acknowledge that starting cold is normal.
- Recommend
/brainstorm opento develop a concept from zero. Mention/brainstorm <hint>if even a single seed word comes to mind. - Show the path:
Concept stage
/brainstorm open— producedesign/concept.md/setup-framework— pin RN / Flutter / iOS / Android/design-bible— visual identity tokens, app icon, splash/map-systems— break the concept into systems/design-system <system>— one PRD per system/consistency-checkand/review-all-prdsArchitecture stage/create-architecture— master architecture/architecture-decision(×N) — ADRs from the required list/create-control-manifest/architecture-reviewPre-production/ux-design— per-screen UX specs/prototype— validate the riskiest mechanic/create-epicsthen/create-stories/sprint-plan— first sprint Production — pick stories with implementation skills (in the second skill set).
If B: Loose theme
- Ask them to share the theme in a sentence or two (free text).
- Validate it as a starting point — never redirect.
- Recommend
/brainstorm <hint>. - Show the same path as A, with the brainstorm step substituted.
If C: Clear pitch
- Ask them to give the one-liner (free text — open response, not
AskUserQuestion). - Use
AskUserQuestion:- Prompt: "How do you want to proceed?"
- Options:
Formalize first— Run/brainstorm <pitch>to structure it intodesign/concept.md.Skip ahead— Go straight to/setup-framework, then author the concept doc by hand.
- Show the path:
/brainstormor/setup-framework(their choice)/design-bibleafter the concept doc exists/design-review design/concept.md/map-systems- then PRDs → architecture → epics as in A.
If D: Existing app
- Reflect what you found in Phase 1, plainly: "I see [X PRDs / Y source files / framework set to Z / no concept doc]."
- Sub-case D1 (early — only a concept or only a framework pin): recommend
/setup-frameworkif missing, then/project-stage-detect. - Sub-case D2 (significant artifacts — PRDs, ADRs, stories already present):
- Explain: "Files existing is not the same as the template's skills being able to read them. PRDs may be missing required sections; ADRs may lack Status.
/adoptchecks compliance specifically." - Recommend in order:
/project-stage-detect— what stage and what is missing entirely/adopt— format compliance and migration plan/setup-frameworkif not pinned/reverse-documentfor any code with no design doc behind it/architecture-reviewto bootstrap the requirements registry
- Explain: "Files existing is not the same as the template's skills being able to read them. PRDs may be missing required sections; ADRs may lack Status.
Phase 3b: Set Review Mode
Check if production/review-mode.txt already exists.
If it exists: Read and report — "Review mode is set to [value]." Move to Phase 4. Do not re-ask.
If it does not exist: Use AskUserQuestion:
- Prompt: "How much review do you want as you move through the workflow?"
- Options:
Full— Director-tier reviews at every gate. Best for teams or anyone learning the template.Lean (recommended)— Reviews only at phase boundaries. Balanced for solo and small teams.Solo— No director reviews. Maximum speed. For prototypes, hackathons, or experienced solo devs.
Write the choice to production/review-mode.txt immediately:
Full→fullLean (recommended)→leanSolo→solo
Create the production/ directory first if it is missing.
Phase 4: Confirm Before Handing Off
Use AskUserQuestion to confirm the recommended first step. Do not run another skill yourself.
- Prompt: "Want to start with [recommended skill]?"
- Options:
Yes — start with [skill]Something else first
If the user picks the alternative, accept their input and adjust.
Phase 5: Hand Off
Respond with one short line: "Type [skill command] to begin." Do not re-explain or pad.
Verdict: COMPLETE — user oriented and routed.
Edge Cases
- User picks D but disk is empty: "The template looks unused — A or B may fit better. Which is closer?"
- User picks A but disk has working code: "I noticed code in [path]. Did you mean D?"
- Returning user, fully set up: Skip the long flow — "Framework: [X]. Concept: present. Review mode: [value or 'lean (default)']. Want to continue with
/sprint-plan,/sprint-status, or describe what you want?" - None of the options fit: Let the user describe their state in free text and adapt.
Examples
User picks B) Loose theme, types "an app that helps bouldering gym climbers track their projects":
- Skill recommends
/brainstorm bouldering project tracker. - Shows the full concept → architecture → production path.
- Asks for review mode (user picks Lean).
- Writes
production/review-mode.txtwithlean. - Confirms first step, hands off.
User picks D) Existing app, project has 3 PRDs and an iOS Xcode project:
- Reports findings.
- Recommends
/project-stage-detectthen/adopt. - Notes that
/setup-frameworkmay not be needed if Xcode metadata already implies iOS native. - Asks for review mode if unset.
- Hands off to
/project-stage-detect.
Collaborative Protocol
- Ask first; never assume the user's state.
- Present options as choices, not orders.
- Recommend, do not auto-execute.
- If the user's situation does not fit a tab, listen and adapt.
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.

