Create Epics
argument-hint: "[system-name | layer: foundation|core|feature|presentation | all] [--review full|lean|solo]"
What Is This
The Create Epics skill (create-epics) is a technical director utility designed for the Happycapy Skills platform. Its primary function is to systematically translate approved Game Design Documents (GDDs) and architectural plans into high-level "epics." Each epic corresponds to a single architectural module, such as a subsystem or a layer (foundation, core, feature, presentation), and sets the boundaries and ownership for that body of work. The skill does not break down epics into implementation tasks or user stories; instead, it prepares the groundwork for subsequent story creation.
This skill enforces strict separation between defining the scope at the module level (epics) and breaking work further into actionable stories. By using well-defined architectural boundaries, it ensures clarity in responsibility, risk tracking, and traceability for the game development process. The output consists of structured documentation files placed within the project’s production/epics/ directory.
Why Use It
The process of building a robust game, especially with a modular architecture, requires clear definition of what each module is, why it exists, and who is responsible for it. The Create Epics skill offers several key benefits:
- Consistent Scope Definition: By mapping each epic to a single architectural module, teams avoid scope creep and maintain clear boundaries.
- Traceability: Epics reference governing Architectural Decision Records (ADRs), engine risks, and any requirements not covered elsewhere.
- Ownership Clarity: Each epic clarifies architectural responsibility, improving accountability.
- Preparedness for Implementation: Epics act as a bridge between high-level design and actionable development tasks, which are created later via stories.
- Layered Development: The skill enforces a disciplined approach, only allowing epics for feature layers once core layers are stable, thereby reducing rework caused by evolving designs.
How to Use It
Invocation Syntax
The skill is invoked using the following format:
/create-epics [system-name | layer: foundation|core|feature|presentation | all] [--review full|lean|solo]system-name: Name of the architectural module or system.layer: Specify the architecture layer (foundation, core, feature, presentation).all: Generate epics for all layers.--review: (Optional) Specify review mode (full,lean, orsolo).
Argument Parsing and Review Mode The skill determines the review mode in this priority order:
- If
--reviewis explicitly provided, use that value. - If not, read from
production/review-mode.txt. - If neither exists, default to
lean.
Example
Suppose you want to generate epics for the core layer with a full review:
/create-epics layer:core --review fullSkill Workflow
- Preconditions: Ensure
/create-control-manifestand/architecture-reviewhave successfully passed. - Execution: For each architectural module within the specified layer, the skill will:
- Create a directory:
production/epics/[epic-slug]/ - Generate an
EPIC.mdfile with:- Epic name and description
- Scope definition
- Governing ADRs
- Engine risk assessment
- Untraced requirements
- Module ownership
- Update or create
production/epics/index.mdto provide an overview of all active epics.
- Create a directory:
Sample EPIC.md Structure
## [Epic Name]
## Description
Brief overview of the architectural module’s purpose and boundaries.
## Scope
- List of included responsibilities
- Explicit exclusions
## Governing ADRs
- ADR-001: Entity-Component System Design
- ADR-003: Serialization Strategy
## Engine Risk
- [Describe any engine-specific risks or uncertainties]
## Untraced Requirements
- [List requirements not covered in GDDs or architecture docs]
## Ownership
- [Team/individual responsible for this module]Post-Epic Workflow After each epic is created, run:
/create-stories [epic-slug]This breaks the epic into actionable development stories.
When to Use It
Use the Create Epics skill under these conditions:
- After the control manifest and architecture review have passed, confirming the design is ready for module breakdown.
- Once you are ready to define the scope and ownership for a specific architecture layer or system.
- When the foundation and core layers are stable, proceed to feature and presentation layers. Do not create epics for dependent layers prematurely, as upstream changes can invalidate designs.
Important Notes
- Strict Layer Order: Only create epics for higher layers (feature, presentation) after lower layers (foundation, core) are nearly complete.
- No Story Breakdown: This skill does not create implementation stories-run
/create-storiesafter each epic is defined. - Review Modes: Choose the review mode appropriate for your workflow. Use
fullfor comprehensive review,leanfor quicker iteration, orsolofor individual use. - Traceability: Ensure that each epic captures all untraced requirements and references relevant ADRs for architectural decisions.
- Documentation Output: All output files are written to
production/epics/, maintaining project documentation hygiene.
By using the Create Epics skill as described, teams ensure a disciplined, traceable, and modular approach to game architecture, setting the stage for efficient and accountable development.
More Skills You Might Like
Explore similar skills to enhance your workflow
Adobe Automation
Automate Adobe operations through Composio's Adobe toolkit via Rube MCP
OpenAPI Explorer
WebFetch https://open.feishu.cn/llms-docs/zh-CN/llms-<module>.txt
Chro Advisor
People leadership for scaling companies. Hiring strategy, compensation design, org structure, culture, and retention. Use when building hiring plans,
Automate Whatsapp
Build WhatsApp automations with Kapso workflows: configure WhatsApp triggers, edit workflow graphs, manage executions, deploy functions, and use
Workflow Orchestration Patterns
- Multi-step processes spanning machines/services/databases
Shadcn UI
Expert guidance for integrating and building applications with shadcn/ui components, including component discovery, installation, customization,