create-epics
Skill40 starsChanged 7 months agoMITView on GitHub
Translates approved PRDs and architecture into epics — one epic per architectural module or major system. Each epic captures scope, governing ADRs, framework risk, and untraced requirements. Run after /architecture-review passes; do NOT decompose into stories — run /create-stories per epic afterward.
10 other projects use a version of this.
Adopted by
Most recently taken up by
- striderZA/OpenCodeGameStudiostheir own version, first seen Nov 4, 2026
- vivekschaudhary/compasstheir own version, first seen Oct 25, 2026
- Donchitos/Claude-Code-Game-Studiostheir own version, first seen Oct 19, 2026
- venus2099/codex-game-studiostheir own version, first seen Oct 17, 2026
- ShiningPlague/shiningplague-studiotheir own version, first seen Oct 11, 2026
- HunLuanZhiZhu/Game-Studios-Plugintheir own version, first seen Oct 5, 2026
Used with
Projects with it also use vertical-slice.
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.
hongyuancSkillvertical-slice
Build and evaluate a production-quality vertical slice before advancing from Pre-Production to Production.
Also in 4 projects that use this
Other versions10 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 10
- Donchitos/Claude-Code-Game-Studios★ 27ksame name
- striderZA/OpenCodeGameStudios★ 118same name
- dawn840705/codex-code-studios★ 17same name
- lilangMax/ClaudeCodeGameStudios★ 8same name
- hongyuanc/codex-game-studios★ 5same name
- vivekschaudhary/compass★ 2same name
- dawn840705/claude-code-studios★ 1same name
- HunLuanZhiZhu/Game-Studios-Plugin★ 1same name
- venus2099/codex-game-studios★ 1same name
- ShiningPlague/shiningplague-studio★ 0same 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 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-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
12 rules. 0 are common, 7 are distinctive, and 5 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Do not decompose into stories here — /create-stories <epic-slug> does that.
- Very specific: line 46
If a previous epic for the same system exists, ask: Refresh, Replace, Cancel
- Very specific: line 48
If <system-name> argument given, scope to that system.
- Distinctive: line 54
Read the inputs in full. Hold:
- Distinctive: line 68
If a system is huge (>30 TR-IDs), split into 2-3 sub-epics (e.g., "Auth — Sign-in", "Auth — Account & Recovery")
- Distinctive: line 184
After all epics are written:
- Very specific: line 185
Update design/systems-index.md to set Status to Epic created for each affected system
- Very specific: line 186
Update production/session-state/active.md with epic count and next-action pointer
- Distinctive: line 204
If any epic has Section 6 entries (untraced requirements), warn loudly:
- Distinctive: line 211
Every TR-ID from every in-scope PRD appears in either Section 5 (covered) or Section 6 (untraced) of some epic
- Distinctive: line 212
Every Governing ADR listed has Status verified — Proposed ADRs trigger BLOCKED labeling
- Distinctive: line 214
No epic has zero TR-IDs (else it is not a real epic — ask the user)
“Common” is measured: kept by at least 6 in 10 of the 10 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
Create Epics
An epic is the unit between architecture and stories. It groups related stories under a shared system or module, declares the governing ADRs, and surfaces the framework risk so the engineer knows what to verify before coding.
Output: production/epics/<epic-slug>/EPIC.md (one file per epic)
Purpose / When to Run
Run when:
/architecture-reviewreturned PASS or CONCERNS (not FAIL)/create-control-manifestis written and dated- The team is ready to plan stories
Do not decompose into stories here — /create-stories <epic-slug> does that.
Inputs
- All
design/prd/*.md - All Accepted
docs/architecture/ADR-*.md docs/architecture/architecture.mddocs/architecture/control-manifest.mddocs/framework-reference/<framework>/- Optional
<system-name>argument: only generate the epic for this system
Outputs
production/epics/<epic-slug>/EPIC.md— one per system in scope
Phase 1: Pre-checks
- Confirm at least one Accepted ADR exists
- Confirm
docs/architecture/control-manifest.mdexists and read its Manifest Version - If a previous epic for the same system exists, ask:
Refresh,Replace,Cancel
If <system-name> argument given, scope to that system; else, do all in-scope MVP systems from the systems index.
Phase 2: Read All Context
Read the inputs in full. Hold:
- Each PRD's Detailed Requirements (all TR-IDs)
- Each PRD's Acceptance Criteria
- Each ADR's PRD Requirements Addressed table
- Architecture Section 3 (Layers) and Section 20 (Required ADRs)
- Control manifest Required/Forbidden/Guardrail by layer
Phase 3: Group TR-IDs into Epics
The mapping rule: one epic per system, with stories living inside.
Special cases:
- If a system is huge (>30 TR-IDs), split into 2-3 sub-epics (e.g., "Auth — Sign-in", "Auth — Account & Recovery")
- Foundation systems (state, navigation, networking) become epics if they need any user-visible scaffolding work
- Cross-cutting concerns (analytics, error reporting) become their own epics if they have ≥3 TR-IDs of dedicated work
Order epics by dependency: Foundation → Core → Features → Polish.
Phase 4: Per-Epic Authoring
For each epic, write production/epics/<slug>/EPIC.md:
# Epic: <System Name>
> **Status**: Ready for stories
> **Layer**: <Foundation / Core / Feature / Polish>
> **Manifest Version**: <date from control-manifest>
> **Created**: <date>
## 1. Goal
<one paragraph from the PRD's Overview + User Goal>
## 2. PRD Source
- `design/prd/<system>.md`
## 3. Governing ADRs
| ADR | Title | Status | Risk |
|-----|-------|--------|------|
| ADR-NNNN | <title> | Accepted | LOW / MED / HIGH |
| ADR-NNNN | <title> | Proposed | (BLOCKED until Accepted) |
## 4. Framework Risk
- Framework: <name + version>
- Knowledge risk: <LOW / MEDIUM / HIGH>
- Verification required (from ADRs): <list>
- Breaking-change touch points: <list or none>
## 5. Requirements Covered
| TR-ID | Requirement summary | Source ADR | Story slug (when created) |
|-------|---------------------|------------|---------------------------|
| TR-<sys>-001 | ... | ADR-NNNN | (pending) |
| ... | | | |
## 6. Untraced Requirements
| TR-ID | Why untraced | Resolution path |
|-------|--------------|-----------------|
| TR-<sys>-007 | No ADR covers this — Proposed only | Run `/architecture-decision` for <area> |
If empty, mark "None — all TR-IDs trace to an Accepted ADR."
## 7. Acceptance for Epic Done
- All TR-IDs in Section 5 implemented and verified
- All stories in this epic at Status: Complete
- All test evidence at the path specified in each story's manifest
- Smoke check passes for this system
- All P0 ADRs covering this epic remain Accepted (no regressions)
## 8. Out of Scope
- <items deliberately excluded — link to where they live (next epic, deferred PRD, v1.1 list)>
## 9. Implementation Notes
### Layer constraints
<from control manifest, copied verbatim — Required and Forbidden for each layer this epic touches>
### Performance budget
<from control manifest guardrails relevant to this epic>
### Test evidence
| Story type expected | Evidence required | Path |
|---------------------|-------------------|------|
| Logic | Automated unit test | tests/unit/<system>/<slug>_test.<ext> |
| Integration | Integration test or documented walkthrough | tests/integration/<system>/<slug>_test.<ext> |
| Visual / Feel | Screenshot + lead sign-off | production/qa/evidence/<slug>-evidence.md |
| UI | Manual walkthrough doc | production/qa/evidence/<slug>-evidence.md |
| Config / Data | Smoke check | production/qa/smoke-<date>.md |
## 10. Risks
- <list — engineering risks that could derail this epic>
## 11. Rollout & Feature Flag
- Feature flag controlling this epic (if any): `<flag_name>` from <provider>
- Rollout plan: <internal → beta → percentage → full>
- Kill switch: <yes / no>
## 12. Open Questions
- <list>
## 13. Story Backlog (filled by `/create-stories`)
[Initially empty. After `/create-stories <slug>`, this lists the story files.]Phase 5: Write & Confirm
Show the draft for the first epic. Ask: "Approve and write to production/epics/<slug>/EPIC.md?"
Loop for each in-scope epic. The user can approve all in batch or one-by-one.
Phase 6: Update Indexes
After all epics are written:
- Update
design/systems-index.mdto set Status toEpic createdfor each affected system - Update
production/session-state/active.mdwith epic count and next-action pointer
Phase 7: Hand Off
Print:
"
<N>epics written. Run/create-stories <epic-slug>for each in dependency order. Foundation epics first."
Use AskUserQuestion:
Run /create-stories for <first foundation epic>Run /sprint-plan(after at least one story batch is created)Stop here
Phase 8: Detect Untraced Requirements
If any epic has Section 6 entries (untraced requirements), warn loudly:
"
<N>requirements across<M>epics are untraced. Stories cannot be implemented safely against untraced requirements. Resolve by writing missing ADRs (/architecture-decision) or marking the requirement out of scope for v1."
Quality Gates
- Every TR-ID from every in-scope PRD appears in either Section 5 (covered) or Section 6 (untraced) of some epic
- Every Governing ADR listed has Status verified — Proposed ADRs trigger BLOCKED labeling
- Manifest Version is recorded at epic creation time
- No epic has zero TR-IDs (else it is not a real epic — ask the user)
Examples
/create-epics
- 8 in-scope systems → 9 epics (auth split into Sign-in and Account)
- 47 TR-IDs covered, 0 untraced
- Foundation epics: navigation-shell, auth-sign-in, persistence
- Output: 9 EPIC.md files
/create-epics auth
- Single-system run. 12 TR-IDs covered by ADR-0003 + ADR-0007. 1 untraced (biometric unlock — no ADR yet).
- Warning surfaces; user chooses to write ADR before story creation.
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.

