Agent skill
sdd-skill-creator
Generate project-specific SDD skills for a repository. Use when: setting up SDD for a new project, customizing SDD for a domain, creating repo-local skills that implement `docs/sdd-guidelines.md`. Triggers: "create sdd skills", "setup sdd for project", "customize sdd"
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/sdd-skill-creator
SKILL.md
SDD Skill Creator
Generate project-specific SDD skills that live in a repository.
Prerequisites
You (the generator) must first:
- Read
docs/sdd-guidelines.md(especially §1 Artifacts, §9 Customization Points, §10 Multi-Agent) - Have project context available
Generated skills will be self-contained—their users won't need to read Guidelines.
Output Structure
{project}/.sdd/skills/
├── sdd-foundation/SKILL.md
├── sdd-requirements/SKILL.md
├── sdd-design/SKILL.md
├── sdd-verify/SKILL.md
└── sdd-workflow/SKILL.md
Instructions
Step 1: Gather Project Context
| Question | Purpose | Default |
|---|---|---|
| Project name? | Naming, examples | — |
| Domain? (game, API, CLI, etc.) | Terminology | — |
| Multi-agent? | Include §10 | No |
| Custom anchor prefixes? | Guidelines §9 | SCOPE-, CONSTRAINT- |
| REQ ID format? | Guidelines §9 | REQ-{NNN} |
Skip questions already answered in conversation.
Step 2: Define Customizations
Fill out the customization record. See templates/customization.yaml.
Step 3: Generate Each Skill
For each skill, create SKILL.md with proper frontmatter and these sections:
| Section | Content |
|---|---|
| Customizations | From Step 2: prefixes, ID formats, terminology |
| Instructions | Guidelines steps rewritten with project terminology |
| Examples | One complete example using project's domain |
| Verification | Checklist to validate output |
Skill-to-Guidelines Mapping
| Skill | Guidelines Source | Key Customizations |
|---|---|---|
| sdd-foundation | §1.1 Foundation | Anchor prefixes, domain framing |
| sdd-requirements | §1.2 Requirements | REQ ID format, alignment examples |
| sdd-design | §1.3 Design, §1.4 Decisions | Rationale patterns, domain design |
| sdd-verify | §3 Verification | Checklists, gap severity |
| sdd-workflow | §1-5 (full flow) | State file, handoff, phase transitions |
What "Project-Adapted" Means
| Guidelines says | You write |
|---|---|
| "Identity Anchors" | "{Domain} Identity Anchors" with project prefixes |
| "REQ-{NNN}" | Project's REQ ID format with example |
| "Design item" | Domain-specific term (e.g., "endpoint", "mechanic") |
If multi-agent: Add §10 coordination to each skill.
Step 4: Write Skills
Create files in .sdd/skills/. Each skill must be:
- Self-contained — User can follow without reading Guidelines
- Project-specific — Uses project terminology throughout
- Guidelines-aligned — Doesn't contradict §1-10
- Concise — Under 200 lines
Step 5: Verify Generated Skills
For each skill:
- Frontmatter has
nameanddescriptionwith triggers? - All Step 2 customization values applied?
- Complete workflow, no placeholders?
- Concrete, domain-relevant example?
- Actionable verification checklist?
Step 6: Inform User
Created SDD skills in {project}/.sdd/skills/:
- sdd-foundation → Foundation documents (§1.1)
- sdd-requirements → Requirements documents (§1.2)
- sdd-design → Design documents (§1.3)
- sdd-verify → Integrity verification (§3)
- sdd-workflow → End-to-end orchestration
Customizations: {prefixes}, {req_format}, multi-agent={yes/no}
Examples
See examples/taskcli.md for a complete walkthrough.
Anti-patterns
| Anti-pattern | Problem | Fix |
|---|---|---|
| Generic skills | No project value | Apply all customizations |
| Contradict Guidelines | Integrity violation | Re-read source section |
| Verbose skills | Context waste | Keep < 200 lines |
| Placeholders left | Incomplete | Fill all {...} before saving |
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?