메뉴
Use when executing implementation plans with independent tasks in the current session
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when starting any conversation — establishes how to find and use Techneering skills, requiring skill invocation before ANY response including clarifying questions
Create distinctive, production-grade frontend interfaces with high design quality. Auto-stacked onto an implementation path when a task involves frontend/UI work — pages, components, layouts, styles, interactions. Also available on demand as tn:craft.
Use after tn:audit passes, when completing major features, or before merging — dispatches a global cross-task code review covering consistency, architecture, and security (distinct from tn:assemble's per-task review)
Verify implementation matches change artifacts — dual-layer verification of code vs spec compliance and code vs real requirements. Use after implementation is complete.
Use when you have a spec or requirements for a multi-step task, before touching code
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
| name | assemble |
| description | Use when executing implementation plans with independent tasks in the current session |
Execute plans by dispatching a fresh subagent per task, with two-stage review after each: spec compliance review first, then code quality review.
FRESH SUBAGENT PER TASK + TWO-STAGE REVIEW = HIGH QUALITY.
Never skip reviews. Never reuse context between tasks.
Use the least powerful model that can handle each role:
Read the implementation plan. Extract all tasks and create a todo list.
For each task:
Frontend Detection: Before dispatching, check if the task involves frontend/UI work. Match against the Frontend Detection Rules in tn:compass (keywords: page/component/UI/layout/style/React/Vue/CSS/Tailwind/form/button/animation etc.). If matched, invoke tn:craft first to get aesthetic guidance, then include that guidance in the implementer prompt.
Use ./implementer-prompt.md template. Provide full task text (never make subagent read the plan file).
One commit per task: the implementer commits its own work before returning (the review stages below compare that commit's diff). Each task = one commit, so 2b/2c reviews have a clean BASE_SHA→HEAD_SHA range per task.
Handle status:
Never ignore an escalation or force the same model to retry without changes.
Use ./spec-reviewer-prompt.md template. Verify implementer built what was requested (nothing more, nothing less).
Use ./code-quality-reviewer-prompt.md template. Only after spec compliance passes.
Mark the task as complete in the todo list and in techneering/changes/<name>/tasks.md.
Analyze task dependencies, build DAG:
| Thought | Reality |
|---|---|
| "Skip review to speed up" | Review is quality assurance, not an optional step |
| "Let the subagent read the plan file" | Must provide full text — don't make subagent read files |
| "Ignore the subagent's BLOCKED" | BLOCKED = needs your help, not "try again" |
| "Run code quality review before spec review" | Order is irreversible: spec first, then quality |
| Dimension | Key point |
|---|---|
| Flow | implementer → spec review → code quality review → next task |
| Subagents | Independent per task, don't inherit session context |
| Parallel | Independent tasks can run parallel, max 3 concurrent |
| Templates | implementer-prompt.md / spec-reviewer-prompt.md / code-quality-reviewer-prompt.md |
| Path placeholder | Always use techneering/changes/<name>/ |
When done:
tn:gate to run test verificationtn:diagnose for systematic debugging