Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/studiokjm/ai-harness-template/decomposegit clone --depth 1 https://github.com/studioKjm/ai-harness-templateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/studiokjm/ai-harness-template/decompose)<a href="https://agentmods.dev/commands/studiokjm/ai-harness-template/decompose"><img src="https://agentmods.dev/badge/commands/studiokjm/ai-harness-template/decompose.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00038 | $0.01118 |
| Opus 5 | $0.00019 | $0.00559 |
| Sonnet 5 | $0.00008 | $0.00224 |
| Haiku 4.5 | $0.00004 | $0.00112 |
Grade A, and why
decompose scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/decompose — Atomic Task Decomposition
/run 전에 태스크를 원자적 단위로 분해하고 검증합니다. 메가 프롬프트를 방지하고, 각 단위가 독립적으로 구현/테스트 가능하도록 합니다.
Instructions
You are the Task Decomposer. Your job is to break down seed spec acceptance criteria into atomic, implementable units.
Why Decompose?
- AI 에이전트는 큰 태스크보다 작은 태스크에서 정확도가 높다
- 원자적 태스크는 독립적으로 테스트 가능하다
- 실패 시 롤백 범위가 작다
- 진행률 추적이 명확하다
Decomposition Rules
1. Atomic Unit Criteria
각 태스크는 다음 조건을 모두 만족해야 합니다:
- 단일 레이어 — 한 태스크가 하나의 레이어(P/L/D)만 터치
- 단일 AC — 한 태스크가 하나의 Acceptance Criteria에 기여
- 독립 테스트 가능 — 다른 태스크 완료 없이도 테스트 가능
- 30분 이내 구현 — 너무 크면 더 분해
- 명확한 완료 기준 — "~가 되면 완료"
2. Decomposition Process
시드 스펙의 각 AC를 분해합니다:
AC-001: "사용자가 검색어를 입력하면 관련 매물을 보여준다"
↓
Task 1 [Data]: 검색 쿼리 레포지토리 메서드 구현 + 테스트
Task 2 [Logic]: 검색 필터링 서비스 (거리 2km, 미판매) + 테스트
Task 3 [Logic]: 검색 결과 정렬/페이지네이션 서비스 + 테스트
Task 4 [Present]: 검색 API 엔드포인트 + 테스트
Task 5 [Present]: 검색 UI 컴포넌트 + 테스트
3. Dependency Ordering
태스크 간 의존성을 명시하고 실행 순서를 결정합니다:
Task 1 (Data)
↓
Task 2 (Logic) ← depends on Task 1
Task 3 (Logic) ← depends on Task 1
↓
Task 4 (Present) ← depends on Task 2, 3
Task 5 (Present) ← depends on Task 4
Validation Checklist
분해 완료 후 검증:
═══ Decomposition Validation ══════════════════════
Total tasks: {N}
By layer: Data: {n} | Logic: {n} | Present: {n}
Dependencies: {dependency graph valid? YES/NO}
All testable: {YES/NO}
All < 30min: {YES/NO}
AC coverage: {all ACs covered? YES/NO}
─────────────────────────────
Result: VALID | NEEDS_REFINE
Anti-patterns to Detect
| Anti-pattern | 증상 | 해결 |
|---|---|---|
| Mega Task | 하나의 태스크가 여러 레이어를 터치 | 레이어별로 분할 |
| Orphan Task | 어떤 AC에도 기여하지 않는 태스크 | 제거 또는 AC에 연결 |
| Circular Dep | 태스크 A→B→A 순환 의존 | 공통 부분을 별도 태스크로 추출 |
| Missing Test | 테스트 작성이 계획에 없음 | 각 태스크에 테스트 포함 |
| Too Small | 한 줄짜리 태스크 | 관련 태스크와 병합 |
Output Format
decomposition:
seed_ref: "seed-v1.yaml"
total_tasks: N
tasks:
- id: "T-001"
ac_ref: "AC-001"
layer: "data"
description: "검색 쿼리 레포지토리 메서드 구현"
test: "검색 조건별 쿼리 결과 검증"
depends_on: []
estimated_minutes: 20
- id: "T-002"
ac_ref: "AC-001"
layer: "logic"
description: "검색 필터링 서비스 구현"
test: "거리/판매상태 필터 로직 단위 테스트"
depends_on: ["T-001"]
estimated_minutes: 25
execution_order:
- phase: 1
tasks: ["T-001"]
- phase: 2
tasks: ["T-002", "T-003"]
- phase: 3
tasks: ["T-004", "T-005"]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 123 lines · 38 tokens per session scan A 3ce30ce114fb
decompose is a command published in the GitHub repository studioKjm/ai-harness-template (43 stars, last pushed 3mo ago), licensed MIT. It adds 38 tokens to every session and 1,118 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
VibeGuard: ExecPlan
Long-term task execution plan — generates self-contained execution documents from SPEC, supports cross-session recovery.
VibeGuard: Review
Structured code review - first run the guard to obtain the baseline, then review according to security → logic → quality → performance priority.
VibeGuard: Live Truth
Verify live claims with fresh facts, inferences, and unresolved gaps.
capture-feedback
Quick feedback capture with structured signals.
check-gates
Run a Pre-Action Gate check against prevention rules before executing a risky action.
harness-adopt
Apply Harness Starter Kit to the current target repository with the prompt-first adoption workflow.