Purpose
This skill provides the complete template library and structural standards for generating comprehensive learning repositories for any SDLC approach (Agile, DDD, Structured Analysis, MBSE, Event-Driven, etc.).
Repository Structure Pattern
{{DIRECTORY_NAME}}/
├── 01-{Phase_1_Name}/
│ ├── 01-{Artifact_1_Name}/
│ │ ├── README.md
│ │ └── examples/
│ │ ├── {example-1}.md
│ │ └── {example-2}.md
│ ├── 02-{Artifact_2_Name}/
│ │ ├── README.md
│ │ └── examples/
│ │ └── {example}.md
│ └── ...
├── 02-{Phase_2_Name}/
│ └── ...
├── README.md
├── {{APPROACH_ABBREV}}_Documents.md
└── tools/
└── link_check.py
Naming Conventions
Phase Folders
- Format:
{NN}-{Phase_Name}/ - NN = 01 to 99 (zero-padded)
- Phase_Name = Descriptive name with underscores (e.g.,
01-Requirements_Gathering)
Artifact Folders
- Format:
{AA}-{Artifact_Name}/ - AA = 01 to 99 (zero-padded, sequential within each phase)
- Artifact_Name = Descriptive name with underscores (e.g.,
01-User_Stories)
Example Files
- Format: Lowercase with hyphens
- Examples:
user-story-template.md,backlog-prioritization.md
Artifact IDs
- Format:
PP-AA - PP = Phase number (01-99)
- AA = Artifact number within phase
- Used for traceability in master tracker
Template: Artifact README.md
# [Artifact Name]
## What this artifact is
[1-3 sentence definition of the artifact in the context of {{APPROACH_NAME}}]
## Intended audience and consumers
[Who creates it, who uses it, who reviews it]
Examples: Product Owner, Scrum Master, Development Team, Stakeholders, Architects, etc.
## Purpose and what it communicates
[Why this artifact exists, what decisions it enables, what problems it solves]
## What it is NOT intended to do
[Clear scope boundaries and anti-patterns to avoid over-engineering]
## When it is mandatory vs. optional (and why)
- **Mandatory when**: [Conditions where this artifact is required]
- **Optional when**: [Conditions where it can be skipped]
- **Rationale**: [Why the governance differs based on context]
## Inputs and upstream dependencies
[What information, artifacts, or activities must happen before this artifact can be created]
Examples:
- Requires: [Artifact ID PP-AA] - [Artifact Name]
- Informed by: [Stakeholder input, market research, etc.]
## Outputs and downstream consumers
[What artifacts, decisions, or activities depend on this artifact]
Examples:
- Feeds into: [Artifact ID PP-AA] - [Artifact Name]
- Enables: [Decision, activity, or workflow]
## Dependencies
[Cross-reference table of related artifacts with IDs]
| Dependency Type | Artifact ID | Artifact Name |
|-----------------|-------------|---------------|
| Prerequisites | PP-AA | [Name] |
| Informs | PP-AA | [Name] |
| Related | PP-AA | [Name] |
## Success criteria and quality checks
[Checklist for "Definition of Done" - when is this artifact complete and usable?]
- [ ] [Criterion 1]
- [ ] [Criterion 2]
- [ ] [Peer review completed]
- [ ] [Stakeholder approval obtained (if required)]
## Examples
See the [examples/](./examples/) folder for:
- [example-1.md](./examples/example-1.md) - [Brief description]
- [example-2.md](./examples/example-2.md) - [Brief description]
Template: Main README.md
# Learn {{APPROACH_NAME}}
A comprehensive, example-driven learning repository for **{{APPROACH_NAME}}** - a {{RIGOR_LEVEL}} rigor, {{FLEXIBILITY_LEVEL}} flexibility approach best suited for **{{BEST_FOR}}**.
## What is {{APPROACH_NAME}}?
[2-4 paragraph introduction to {{APPROACH_NAME}}]
- Core philosophy and principles
- When to use this approach
- Key differentiators from other approaches
- Reference to {{MAIN_REFERENCE_URL}}
## Repository Structure
This repository contains **{{TOTAL_ARTIFACTS}} artifacts** organized across **{{NUM_PHASES}} lifecycle phases**:
1. **Phase 01: {Phase_1_Name}** - [Brief description]
2. **Phase 02: {Phase_2_Name}** - [Brief description]
...
{{NUM_PHASES}}. **Phase {{NUM_PHASES}}: {Phase_N_Name}** - [Brief description]
Each artifact includes:
- **README.md**: Specification, purpose, audience, dependencies, success criteria
- **examples/**: Realistic examples using the **{{DOMAIN_EXAMPLE}}** domain
## How to Use This Repository
### For Learners
1. Start with [{{APPROACH_ABBREV}}_Documents.md](./{{APPROACH_ABBREV}}_Documents.md) for a complete artifact index
2. Follow the phases sequentially or jump to specific artifacts
3. Read the README for context, then study the examples
4. Trace dependencies to understand artifact relationships
### For Practitioners
- Use as a reference for deliverable templates
- Adapt examples to your own domain
- Use success criteria as quality gates
- Leverage dependency matrix for project planning
### For Teams
- Establish which artifacts are mandatory vs. optional for your project
- Customize examples to your specific context
- Use as onboarding material for new team members
## Example Domain: {{DOMAIN_EXAMPLE}}
All examples use a consistent **{{DOMAIN_EXAMPLE}}** domain to demonstrate cross-artifact traceability. This allows you to see how a single requirement flows through multiple lifecycle phases.
## Quality Validation
Run `tools/link_check.py` to validate:
- All internal links in {{APPROACH_ABBREV}}_Documents.md resolve correctly
- All example files exist and are accessible
- Cross-references between artifacts are valid
```bash
python tools/link_check.py
Contributing
To add new artifacts or improve examples:
- Follow the naming conventions in this README
- Use the artifact README template from
.claude/skills/sdlc-repository-templates - Ensure examples use the {{DOMAIN_EXAMPLE}} domain for consistency
- Update {{APPROACH_ABBREV}}_Documents.md with new entries
- Run link validation before submitting
References
- Official Source: {{MAIN_REFERENCE_URL}}
- Rigor Level: {{RIGOR_LEVEL}}
- Flexibility: {{FLEXIBILITY_LEVEL}}
- Best For: {{BEST_FOR}}
License
[Specify license - e.g., MIT, CC BY 4.0, etc.]
## Template: Master Tracker (_Documents.md)
```markdown
# {{APPROACH_NAME}} - Master Artifact Tracker
This document provides a complete index of all {{APPROACH_NAME}} artifacts organized by lifecycle phase.
## Quick Navigation
- [Phase 01: {Phase_1_Name}](#phase-01-phase_1_name)
- [Phase 02: {Phase_2_Name}](#phase-02-phase_2_name)
...
## Artifact Overview Table
| ID | Phase | Artifact | Folder | README | Example | Dependencies/Inputs |
|----|-------|----------|--------|--------|---------|---------------------|
| 01-01 | {Phase_1_Name} | {Artifact_1} | [01-{Phase_1}/01-{Artifact_1}/](./01-{Phase_1}/01-{Artifact_1}/) | ✅ | ✅ | — |
| 01-02 | {Phase_1_Name} | {Artifact_2} | [01-{Phase_1}/01-{Artifact_2}/](./01-{Phase_1}/01-{Artifact_2}/) | ✅ | ✅ | 01-01 |
## Phase Details
### Phase 01: {Phase_1_Name}
**Purpose**: [What this phase achieves in {{APPROACH_NAME}}]
**Artifacts**:
1. **[01-01] {Artifact_1_Name}** - [One-line description]
- Examples: {example-1.md}, {example-2.md}
- Dependencies: None
2. **[01-02] {Artifact_2_Name}** - [One-line description]
- Examples: {example.md}
- Dependencies: 01-01
## Dependency Matrix
| Artifact ID | Depends On | Feeds Into |
|-------------|------------|------------|
| 01-01 | — | 01-02, 02-01 |
| 01-02 | 01-01 | 02-03 |
## Completion Status
- **Total Artifacts**: {{TOTAL_ARTIFACTS}}
- **READMEs Complete**: {{README_COUNT}}/{{TOTAL_ARTIFACTS}}
- **Examples Complete**: {{EXAMPLE_COUNT}}/{{TOTAL_ARTIFACTS}}
Phase Definition Guidelines by Approach Type
For Agile/Iterative
Phases: Discovery, Sprint Planning, Development Iteration, Review & Retrospective, Release, etc. Artifacts: User Stories, Product Backlog, Sprint Backlog, Definition of Done, Burndown Charts
For DDD (Domain-Driven Design)
Phases: Strategic Design, Bounded Context Mapping, Tactical Design, Domain Events, Implementation Artifacts: Context Map, Ubiquitous Language, Aggregate Design, Domain Events Catalog
For Structured Analysis
Phases: Requirements, Analysis, Design, Implementation, Testing, Deployment, Operations Artifacts: DFDs, Data Dictionary, Process Specs, ERDs, Structure Charts
For MBSE (Model-Based Systems Engineering)
Phases: Stakeholder Needs, Requirements Analysis, Logical Design, Physical Design, Verification Artifacts: SysML Diagrams, Requirements Models, Interface Definitions
For Event-Driven
Phases: Event Storming, Event Catalog, Schema Design, Consumer Design, Implementation Artifacts: Event Map, Event Schemas, Consumer Contracts, Event Flows
Example File Requirements
Domain Consistency
- Use {{DOMAIN_EXAMPLE}} as the primary domain across all artifacts
- Creates cross-cutting comprehension (e.g., same user stories feed into same sprint plans)
Realistic Scale
- Examples should be pedagogical (20-80 lines), not production-scale
- Show enough detail to be instructive
- Avoid overwhelming learners with complexity
Traceability
Examples should reference:
- Upstream artifact IDs (e.g., "Based on User Story US-042 from 02-01")
- Requirement IDs, feature IDs, or other tracking identifiers
- Downstream consumers (e.g., "Feeds into Sprint Backlog 03-02")
File Formats
- Primarily Markdown (.md) for textual artifacts
- Mermaid diagrams (.mmd or inline in .md) for visual models
- JSON/YAML for configuration examples
- Code snippets (.py, .js, .sql) where appropriate
- Screenshots or mockups (PNG/SVG) for UI artifacts
Quality Criteria Checklist
✅ Structure Consistency: All phases and artifacts follow naming conventions ✅ Template Adherence: All READMEs use the standardized template ✅ Domain Consistency: All examples use {{DOMAIN_EXAMPLE}} domain ✅ Traceability: Dependencies clearly documented in master tracker ✅ Completeness: Every artifact has README + at least 1 example ✅ Validation: Link checker passes without errors ✅ Alignment: Content reflects {{RIGOR_LEVEL}} rigor and {{FLEXIBILITY_LEVEL}} flexibility ✅ Authority: Artifacts reference best practices from {{MAIN_REFERENCE_URL}}
Best Practices
When Creating Phases
- Align with approach philosophy and methodology
- Consider rigor level (more formal = more phases)
- Consider flexibility (adaptive approaches may combine phases)
- Typically 8-14 phases for comprehensive coverage
When Defining Artifacts
- Core deliverables only (3-11 per phase)
- Industry-standard artifacts from authoritative sources
- Appropriate to rigor level (formal approaches have more documentation)
- Aligned with flexibility (adaptive approaches have lighter artifacts)
When Writing Examples
- Use consistent domain throughout repository
- Include traceability IDs
- Show realistic but not overwhelming detail
- Reference upstream and downstream artifacts
- Use proper file formats (markdown, mermaid, code)
Common Pitfalls to Avoid
❌ Inconsistent Naming: Mixing underscores/hyphens, wrong padding ❌ Template Deviation: Skipping sections or using custom formats ❌ Domain Switching: Using different example domains across artifacts ❌ Missing Traceability: Not linking artifacts with IDs ❌ Over-Engineering: Production-scale examples that overwhelm learners ❌ Under-Specification: Examples too trivial to be instructive
Related Skills
testing-conventions: For creating test-related artifactsapi-standards: For API design artifactsdeployment-process: For deployment and operations artifacts