adopt
Skill40 starsChanged 7 months agoMITView on GitHub
Brownfield onboarding for an existing mobile-app codebase. Audits PRDs, ADRs, stories, UX specs, and architecture docs for compatibility with this template's review skills, classifies gaps by impact, and produces a numbered migration plan. Use when joining a project mid-flight or upgrading from an older template version.
27 other projects use a version of this.
Adopted by
Most recently taken up by
- maideps/ai-harness-frameworktheir own version, first seen Nov 8, 2026
- striderZA/OpenCodeGameStudiostheir own version, first seen Nov 4, 2026
- myowinthein/claude-helmtheir own version, first seen Nov 1, 2026
- dcbruck/claude-skillstheir own version, first seen Oct 20, 2026
- iadr-dev/greenlighttheir own version, first seen Oct 20, 2026
- HurricaHjz/SecondSelftheir own version, first seen Oct 20, 2026
Other versions27 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 27
- Donchitos/Claude-Code-Game-Studios★ 27ksame name
- HurricaHjz/SecondSelf★ 156same name
- striderZA/OpenCodeGameStudios★ 118same name
- HurricaHjz/second-yourself★ 105same name
- mrveiss/AutoBot-AI★ 39same name
- dawn840705/codex-code-studios★ 17same name
- lilangMax/ClaudeCodeGameStudios★ 8same name
- hongyuanc/codex-game-studios★ 5same name
- codebycorey/delta-spec★ 4same name
- gonimar/claude-web-studio★ 4same name
17 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
- 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
- startSkill
- story-doneSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 25 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
All design, architecture, production, and ux artifacts on disk
- Very specific: line 160
If the user said yes, write production/migration-plan.md:
- Very specific: line 196
Ask: "May I write this plan to production/migration-plan.md?"
- Very specific: line 205
If framework was a HIGH finding: insist that /setup-framework runs first, since several review skills rely on the framework reference docs.
- Distinctive: line 213
If no findings are CRITICAL or HIGH, output "Project is compatible with all template skills." and skip writing the plan.
“Common” is measured: kept by at least 6 in 10 of the 25 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
Adopt — Brownfield Onboarding
Where /project-stage-detect answers "what exists?", this skill answers "will what exists actually work with /design-review, /architecture-review, /story-readiness, etc.?" Files can be present and still fail compliance because they lack the structured sections those reviewers expect.
This skill writes one file: production/migration-plan.md. The user runs the migration steps separately.
Purpose / When to Run
Run when:
- Pulling into an existing mobile-app project for the first time
- Migrating from an earlier version of this template
- The user has many artifacts but several review skills throw "missing section" errors
- Preparing to hand the project to a new contributor and wanting to know what is broken
Distinct from /project-stage-detect (existence check) and /onboard (per-role context summary).
Inputs
- All design, architecture, production, and ux artifacts on disk
.claude/docs/technical-preferences.md(framework pin)
Outputs
production/migration-plan.md— numbered, prioritized list of fixes- A printed summary
Phase 1: Inventory
Glob and read (sample, not full):
design/concept.mddesign/design-bible.mddesign/prd/*.mddesign/ux/*.mddesign/systems-index.mddocs/architecture/architecture.mddocs/architecture/ADR-*.mddocs/architecture/control-manifest.mddocs/architecture/tr-registry.yamlproduction/epics/*/EPIC.mdproduction/epics/*/story-*.md
Count each. Print the inventory before doing the audit.
Phase 2: Compliance Audit
For each artifact type, check against the template's required structure. Compliance is per-section — partial credit is allowed. Classify gaps as:
- CRITICAL — review skills will crash or refuse to run
- HIGH — review skills will run but produce useless output (missing traceability)
- MEDIUM — degraded but usable
- LOW — cosmetic
2a: Concept doc (design/concept.md)
Required sections (grep): One-liner, Target User, Core Promise, MVP Scope, Primary Metric, Tone, Monetization Stance, Riskiest Assumption.
- Missing One-liner / Target User / Primary Metric: HIGH
- Missing other sections: MEDIUM
2b: PRDs (design/prd/*.md)
Required sections per PRD (grep): Overview, User Goal, Detailed Requirements, Edge Cases, Dependencies, Configurable Values, Acceptance Criteria, Accessibility.
- Missing Acceptance Criteria: CRITICAL (
/prd-reviewand/create-storiescannot run) - Missing Edge Cases / Dependencies: HIGH
- Missing Accessibility: HIGH for any UI-touching system, MEDIUM otherwise
- Each requirement should have a
TR-<system>-NNNID: missing IDs → HIGH (traceability lost)
2c: UX specs (design/ux/*.md)
Required sections: Screen Goal, User Flow, Layout, States, Empty / Loading / Error, Accessibility, Platform Notes, Implementation Hooks.
- Missing States or Empty/Loading/Error: HIGH (mobile apps live in these states)
- Missing Accessibility: HIGH (WCAG 2.2 AA + platform conventions are template requirements)
2d: Architecture (docs/architecture/)
architecture.mdexists with sections:Overview,Layers,Cross-cutting Concerns,Data Flow,Required ADR List. Missing Required ADR List: HIGH.- Each
ADR-*.mdmust have:Status,Date,Framework Compatibility,ADR Dependencies,Context,Decision,Alternatives,Consequences,PRD Requirements Addressed. Missing Status: CRITICAL. Missing Framework Compatibility: HIGH (knowledge-cutoff risk unknown). Missing PRD Requirements Addressed: HIGH (traceability lost). control-manifest.mdexists: if missing and ≥3 Accepted ADRs exist, MEDIUM.tr-registry.yamlexists: if missing and stories exist, HIGH.
2e: Stories (production/epics/*/story-*.md)
Required for each story:
> **Status**:line> **Type**:line (Logic / Integration / Visual / UI / Config)- Embedded TR-ID reference
- Embedded ADR reference (or explicit "no ADR — quick design embedded")
## Acceptance Criteria## Test Evidence Path
Missing Status or Acceptance Criteria: CRITICAL.
Missing Type: HIGH (/story-done cannot pick the right gate).
Missing TR-ID or ADR: HIGH (drift risk).
2f: Framework reference
.claude/docs/technical-preferences.mdFramework field still[TO BE CONFIGURED]AND code clearly uses a framework: HIGH.docs/framework-reference/<name>/VERSION.mdmissing while code uses that framework: MEDIUM.
Phase 3: Classify and Group
Group findings by artifact type and severity. Within each severity tier, sort by:
- Files referenced by other artifacts (a broken upstream PRD blocks downstream stories — fix first)
- Files most-referenced by review skills
For each finding, record:
- File path
- Severity (CRITICAL / HIGH / MEDIUM / LOW)
- Specific gap (e.g., "missing
## Acceptance Criteriaheading") - Suggested fix skill (e.g.,
/design-system retrofit design/prd/auth.md)
Phase 4: Present Findings & Confirm Scope
Show the user a summary table:
## Compliance Audit Summary
| Artifact | Total | Compliant | Critical | High | Medium |
|----------|-------|-----------|----------|------|--------|
| Concept doc | 1 | 0 | 0 | 1 | 0 |
| PRDs | 5 | 2 | 1 | 2 | 0 |
| UX specs | 8 | 4 | 0 | 3 | 1 |
| ADRs | 4 | 1 | 1 | 2 | 0 |
| Stories | 23 | 18 | 2 | 3 | 0 |
| Framework ref | 0 | 0 | 0 | 1 | 0 |Then list every CRITICAL and HIGH issue with file path + specific gap + recommended fix.
Use AskUserQuestion:
- Prompt: "Want a written migration plan covering these fixes?"
- Options:
Yes — full plan (CRITICAL + HIGH + MEDIUM)Yes — minimal plan (CRITICAL + HIGH only)No — just show me the summary
Phase 5: Write Migration Plan
If the user said yes, write production/migration-plan.md:
# Migration Plan
> Generated: [date]
> Scope: [Full / Minimal]
> Total fixes: [N]
## Phase A — Unblock review skills (CRITICAL)
Fixes here let `/prd-review`, `/architecture-review`, `/story-readiness`, and `/story-done` actually run.
1. **[file path]** — [gap]
- Fix: `/design-system retrofit [path]` (or specific manual step)
- Why: `/prd-review` will fail without `## Acceptance Criteria`.
[... numbered list ...]
## Phase B — Restore traceability (HIGH)
Fixes here let cross-document review (`/architecture-review`, `/review-all-prds`) produce useful output.
[... numbered list ...]
## Phase C — Polish (MEDIUM)
Quality-of-life fixes — not blocking.
[... numbered list ...]
## Recommended order
1. Run all of Phase A first.
2. After Phase A is clear, run `/review-all-prds` and `/architecture-review` to verify the fixes worked.
3. Then proceed with Phase B.
4. Phase C can be done lazily during normal work.Ask: "May I write this plan to production/migration-plan.md?"
Phase 6: Hand Off
After writing:
- Print: "Migration plan written. Start with Phase A item 1: run
[first fix command]." - Suggest running
/project-stage-detectafter Phase A is complete to confirm stage classification has improved. - If framework was a HIGH finding: insist that
/setup-frameworkruns first, since several review skills rely on the framework reference docs.
Quality Gates
- The plan must list a specific fix command for every CRITICAL and HIGH item — never just "fix this manually" with no path.
- The plan must order Phase A by upstream-first (PRDs before stories that reference them).
- If no findings are CRITICAL or HIGH, output "Project is compatible with all template skills." and skip writing the plan.
Examples
Project has 5 PRDs; one is missing ## Acceptance Criteria. 23 stories; 2 have no Status line. ADRs all present but one lacks Framework Compatibility.
Audit verdict:
- 1 CRITICAL (missing Acceptance Criteria —
/prd-reviewwill fail) - 1 CRITICAL (2 stories missing Status —
/story-donecannot resolve them) - 1 HIGH (ADR missing Framework Compatibility — knowledge-gap risk untracked)
Migration plan:
- Phase A.1:
/design-system retrofit design/prd/billing.md - Phase A.2: manually add
> **Status**: Readyto the two stories - Phase B.1:
/architecture-decision retrofit docs/architecture/ADR-0003-payment.md
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.

