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.
git clone --depth 1 https://github.com/jed1978/ddd-architecture-coachWrote 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/jed1978/ddd-architecture-coach/phase-4)<a href="https://agentmods.dev/commands/jed1978/ddd-architecture-coach/phase-4"><img src="https://agentmods.dev/badge/commands/jed1978/ddd-architecture-coach/phase-4.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 | $0.00072 | $0.00382 |
| Opus 5 | $0.00036 | $0.00191 |
| Sonnet 5 | $0.00014 | $0.00076 |
| Haiku 4.5 | $0.00007 | $0.00038 |
Grade A, and why
phase-4 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 6d 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.
What it actually says
請啟動 DDD Architecture Coach（ddd-architecture-coach skill）並強制進入 Phase 4。
Argument handling:
$ARGUMENTS 為 review 範圍（可選）：
- 空 → full review（system + 所有 BC）
Phase 1/Phase 2→ 對應 phase 的 system-level reviewPhase 3:<BC>或<BC>→ 該 BC 的 Phase 3 reviewsystem→ 僅 system 層 artifactsfull→ 等同於空（明確版）
執行步驟：
- 讀取
.claude/arch-learnings.md（避免重複 flag 已解決的問題）；.claude/arch-state.md僅在 review 結論為 rollback retarget 時更新 - 讀取
references/phase4-review-iterate.md - 依 $ARGUMENTS 判斷 review scope
- 執行 Pre-flight Checks，依適用的 checklist 產出評分報告
- 若 review 結論為 rollback：更新
.claude/arch-state.md的last_touched: { bc: <retarget BC 或空>, phase: <retarget phase>, at: <today> }；否則不動arch-state.md
$ARGUMENTS
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.
- 6d ago First seen · 24 lines · 72 tokens per session scan A b2a4cc491a07
phase-4 is a command published in the GitHub repository jed1978/ddd-architecture-coach (5 stars, last pushed 4mo ago), licensed MIT. It adds 72 tokens to every session and 382 once invoked, about $0.0004 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 2026-08-31.
Other commands, from other repositories
tenets-scaffold
You are initializing the architecture foundation for the Flask service in the current repository. This workflow supports one service only. It does not generate individual bounded contexts, workflows, features, or resources.
tenets-review-architecture
You are a strict architecture reviewer for a codebase following Hexagonal Architecture (Ports & Adapters) with Domain-Driven Design.
review
Submit a story for verification (In Progress → In Review, no PR yet).
wp-cleanup
Find and clean up unused PHP code, Composer dependencies, imports, and dead code in WordPress projects.
audit
Adversarial, exhaustive review of the entire repository across code, tests, documentation, and configuration, with optional fixing in the same run. Assumes the code is incorrect until proven otherwise. This is the default command when the invocation names no other.
a
Intercept Mode - Auto-improve and execute immediately.