design
AIDLC Design phase — Tech Spec under feature/<slug>/, review passes, human gate before /build.
Skill by queen-of-code
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/queen-of-code-design
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/design
What's inside
- Skill name
design- Description
- AIDLC Design phase — Tech Spec under feature/<slug>/, review passes, human gate before /build. Requires an approved Product Spec (run /plan first or confirm approval in-thread).
- License
- MIT
- Instructions
- 43 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# /design — Design (Tech Spec) You are the **phase orchestrator** for AIDLC **Design** (Tech Spec). Ground truth is **`docs/AIDLC.md`** in the **consumer workspace**. **Plan (Product Spec)** is **`/plan`** ([skills/plan/SKILL.md](../plan/SKILL.md)). This skill assumes **Product Spec is approved** (or the user explicitly approves proceeding in the current thread). **Library skills** — [docs/SKILLS.md](../../docs/SKILLS.md); resolve from your install or `.claude/skills/<bundle>/`. ## Before you start 1. Resolve **feature slug** from `$ARGUMENTS` or from the same folder the team used for `/plan`. 2. **Read `feature/<slug>/product-spec.md`**. If it is missing or clearly not approved, **stop** and ask the human to run **`/plan`** or confirm approval — do not invent product scope. 3. Ensure `feature/<slug>/` exists. If `tech-spec.md` is missing, create it from **[`tech-spec-template.md`](../spec-management/templates/tech-spec-template.md)**. 4. Link the work to the same **parent work item** as `/plan` (per **`AGENTS.md` → Issue tracker (AIDLC)**). For GitHub-only automation, see [GITHUB-AIDLC-PROJECT.md](https://github.com/queen-of-code/AI-DLC/blob/main/docs/GITHUB-AIDLC-PROJECT.md). Portability: [ISSUE-TRACKER-PORTABILITY.md](https://github.com/queen-of-code/AI-DLC/blob/main/docs/ISSUE-TRACKER-PORTABILITY.md). ## Orchestration — Tech Spec (`tech-spec.md`) 1. Translate the **approved** Product Spec into one or more **Units**; one `tech-spec.md` in this feature folder unless work is split across sub-issues (link related specs in the doc). 2. Include: scope, architecture, API/UI contracts, data model, acceptance criteria for **Review**, **testing approach** (what Build+Test must cover), risks — per AIDLC Design in `docs/AIDLC.md`. - **Architecturally-relevant changes** — anything touching a **state/status field, an enum, an async handoff, a gated or terminal action, a multi-step pipeline, or a publish/compile step** (trivial copy/CSS/config is exempt) — **must** additionally carry, per **[docs/ARCHITECTURAL-SOUNDNESS.md](../../docs/ARCHITECTURAL-SOUNDNESS.md)**: a **state machine** (states + a single transition function every path routes through + why illegal states are unreachable *by construction*), **sequence diagram(s)** for each cross-component/async handoff, the **invariants + their single enforcement chokepoint**, and the **validity boundary** (where invalid states are made unreachable — prefer schema/type/compile/publish over run-time guards). Without these, the change is **not ready** for the Design gate; `/review`'s Architectural Soundness pass scores against them. 3. **Tech Spec review passes** (run in order; merge findings into the doc; open issues in an appendix if needed): | Pass | Library skill | |------|----------------| | Architecture / boundaries | `architecture` ([skills/architecture/SKILL.md](../architecture/SKILL.md)) | | Frontend | `frontend-web` ([skills/frontend-web/SKILL.md](../frontend-web/SKILL.md)) | | Backend / API | `backend-saas` ([skills/backend-saas/SKILL.md](../backend-saas/SKILL.md)) | | Testing strategy | `testing` ([skills/testing/SKILL.md](../testing/SKILL.md)) | | CI / Docker / deploy | `architecture` + read `.github/workflows/`, `docker-compose`, Dockerfiles | 4. **Stop for human approval** of the Tech Spec before **`/build`**. ## Outputs - `feature/<slug>/tech-spec.md` (and linked ADR drafts under `adr/` if your repo uses them per **spec-management**) ## Rules - Do not reopen settled Product decisions in the Tech Spec without flagging a **change request** to Product. - **Conversation first** for technical ambiguities — same rhythm as `docs/AIDLC.md` orchestration model. **Headless:** ask on the work item with the owner @mentioned and **halt** until answered ([ASK-AND-HALT.md](../../docs/ASK-AND-HALT.md)); the same applies to the "stop and ask" in *Before you start*. - Mark any example that must be reproduced exactly as **exact**; unmarked examples are illustrative ([INTENT-OVER-LITERAL.md](../../docs/INTENT-OVER-LITERAL.md)).
From queen-of-code/AI-DLC, MIT.
More from queen-of-code/AI-DLC
- 0.625score
agent-product-manager
queen-of-code
Writes feature spec drafts using the product template.
SkillAFreshness: A★ 9Claude Code +9 - 0.625score
build
queen-of-code
AIDLC Build + Test orchestrator (TDD).
SkillAFreshness: A★ 9Claude Code +9 - 0.595score
agent-reviewer
queen-of-code
Review Orchestrator.
SkillAFreshness: A★ 9Claude Code +9 - 0.538score
frontend-web
queen-of-code
Modern web development patterns for React, Vue, and vanilla JS including component architecture, state management, and performance.
SkillCFreshness: C★ 9Claude Code +9 - 0.535score
agent-issue-tracker-setup
queen-of-code
Onboards AIDLC issue-tracker choice for a consumer repo: fills AGENTS.md Issue tracker (AIDLC) table, checklists for GitHub vs Linear vs Jira paths, and links…
SkillCFreshness: C★ 9Claude Code +9 - 0.473score
backend-saas
queen-of-code
SaaS backend development patterns including API design, multi-tenancy, authentication, and scalability.
SkillDFreshness: D★ 9Claude Code +9 - 0.473score
architecture
queen-of-code
Apply software architecture best practices and design patterns.
SkillDFreshness: D★ 9Claude Code +9 - 0.473score
blog-writing
queen-of-code
Write blog posts in Melissa Benua's voice and style, saved to external-brain/blog-posts.
SkillDFreshness: D★ 9Claude Code +9 - 0.470score
agent-learn
queen-of-code
Captures learnings after a feature or incident completes.
SkillDFreshness: D★ 9Claude Code +9 - 0.470score
agent-devops-review
queen-of-code
DevOps review pass for PRs — CI/CD, containers, workflows, rollout, rollback, monitoring vs Tech Spec.
SkillDFreshness: D★ 9Claude Code +9 - 0.470score
agent-security-review
queen-of-code
Lightweight security review pass for PRs — secrets, auth/access, dependencies, web/data safety, config.
SkillDFreshness: D★ 9Claude Code +9 - 0.440score
agent-researcher
queen-of-code
Explores a codebase, documentation set, brain-mcp collection, or external sources to answer a specific question.
SkillDFreshness: D★ 9Claude Code +9
Signals
- Stars
- 9
- Forks
- 13
- Open issues
- 0
- Last commit (this folder)
- Sep 25, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/queen-of-code/AI-DLC
- Website
- queenofcode.dev
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
- 0.283score
design
csackrider
AIDLC Design phase — Tech Spec under feature/<slug>/, review passes, human gate before /build.
SkillCFreshness: CClaude Code +9 - 0.244score
design
nathan-oakes
AIDLC Design phase — Tech Spec under feature/<slug>/, review passes, human gate before /build.
SkillDFreshness: DClaude Code +9
More from queen-of-code
- 0.625score
build
queen-of-code
AIDLC Build + Test orchestrator (TDD).
SkillAFreshness: A★ 9Claude Code +9 - 0.625score
agent-product-manager
queen-of-code
Writes feature spec drafts using the product template.
SkillAFreshness: A★ 9Claude Code +9 - 0.595score
agent-reviewer
queen-of-code
Review Orchestrator.
SkillAFreshness: A★ 9Claude Code +9 - 0.538score
frontend-web
queen-of-code
Modern web development patterns for React, Vue, and vanilla JS including component architecture, state management, and performance.
SkillCFreshness: C★ 9Claude Code +9 - 0.535score
agent-issue-tracker-setup
queen-of-code
Onboards AIDLC issue-tracker choice for a consumer repo: fills AGENTS.md Issue tracker (AIDLC) table, checklists for GitHub vs Linear vs Jira paths, and links…
SkillCFreshness: C★ 9Claude Code +9 - 0.473score
architecture
queen-of-code
Apply software architecture best practices and design patterns.
SkillDFreshness: D★ 9Claude Code +9
Cite this page
design skill. Agent Mart. https://agentmart.dev/skills/queen-of-code-design (last updated 2026-10-24)