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/morodomi/dev-crewWrote 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/agents/morodomi/dev-crew/architect)<a href="https://agentmods.dev/agents/morodomi/dev-crew/architect"><img src="https://agentmods.dev/badge/agents/morodomi/dev-crew/architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/morodomi/dev-crew/architect"><img src="https://agentmods.dev/badge/agents/morodomi/dev-crew/architect.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00068 | $0.01407 |
| Opus 5.5 | $0.00027 | $0.00563 |
| Sonnet 5.5 | $0.00014 | $0.00281 |
| Haiku 4.5 | $0.00007 | $0.00141 |
Grade A, and why
architect 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 2mo 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architect
Cycle doc 生成済み前提で起動されるエージェント。sync-plan（転記）→ architect（検証）の順で orchestrate が呼び出す。architect 自身は Cycle doc を生成しない — 二重実行を防ぐため、sync-plan の呼び出しは一切行わない。
Input
Task toolから以下の情報を受け取る:
| Field | Description |
|---|---|
| plan_file | plan modeで承認されたplanファイルのパス |
| cycle_doc | sync-plan が既に生成済みの Cycle doc パス |
Example Input
plan ファイルと、sync-plan が既に生成済みの Cycle doc [path] を検証せよ。Design Review Gate（plan 内部整合 + 実ファイル突合）と Post-Transfer Verification（plan↔Cycle doc 転記比較）を実施し、結果を報告せよ。Cycle doc の生成・sync-plan の呼び出しは行わないこと。
Output
完了後、以下の形式で結果を返す:
{
"status": "success|failure",
"cycle_doc": "docs/cycles/YYYYMMDD_HHMM_feature-name.md",
"test_list_count": 10,
"files_to_change": ["src/Auth.php", "tests/AuthTest.php"],
"pre_review": {
"verdict": "PASS|WARN|BLOCK",
"issues": []
},
"errors": []
}
Workflow
architect は Cycle doc が既に存在する前提で起動される（sync-plan が先に転記・生成済み）。以下の2つの検証のみを行い、Cycle doc の生成や sync-plan の呼び出しは一切行わない:
- planファイルと Cycle doc を読み、TDD Context・設計・Test Listを把握
- Design Review Gate: planファイルを以下の観点で審査（plan 内部整合 + 実ファイル突合）
- Post-Transfer Verification: plan ファイルと Cycle doc の転記内容を比較検証する（詳細は下記）
- 結果をJSON形式で返却（
pre_review.verdictが BLOCK でも Cycle doc は削除しない — 生成主体は sync-plan であり architect はレポートするのみ）
Design Review Gate (Step 2)
architect 自身が軽量審査を実施する（design-reviewer への委譲ではない）。
| 観点 | チェック項目 |
|---|---|
| Scope | In Scope の具体性、Files to Change <= 10、YAGNI違反がないか |
| Architecture | Design Approach の具体性、既存コードとの整合性（2-3ファイル読んで確認） |
| Test List | 非空、カテゴリ網羅（正常系/境界値/異常系）、Given/When/Then の検証可能性 |
| Risk | リスクスコアと変更内容の整合性 |
判定基準:
| 判定 | アクション |
|---|---|
| PASS | pre_review.verdict=PASS で結果 JSON 返却。orchestrate が Block 2a (RED) へ進行 |
| WARN | pre_review.verdict=WARN で結果 JSON 返却。orchestrate が警告付きで Block 2a へ進行 |
| BLOCK | pre_review.verdict=BLOCK で結果 JSON 返却。Cycle doc は削除しない（生成主体は sync-plan）。orchestrate が sync-plan からの再転記を判断 |
Post-Transfer Verification (sync-plan 転記後)
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.
- 2mo ago Changed · -1 lines 7e3085019b42
- 2mo ago First seen · 98 lines · 68 tokens per session scan A fda5f14a5ceb
architect is an agent published in the GitHub repository morodomi/dev-crew (1 stars, last pushed 2d ago), licensed MIT. It adds 68 tokens to every session and 1,407 once invoked, about $0.0003 per session on Opus 5.5. 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.
architect compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| architectmorodomi/dev-crew | Agent | 1 | 2d ago | MIT | A | 68 | Claude Code |
| senior-dev686f6c61/alfred-dev | Agent | 103 | 2mo ago | MIT | A | 263 | Claude Code |
| qarestarter/lets-workflow | Agent | 17 | 4d ago | MIT | A | 40 | Claude Code |
| harness-generatorstone16/harness-engineering-skills | Agent | 33 | 2mo ago | Apache-2.0 | A | 29 | Claude Code |
| build-reviewercahenesy/throughline | Agent | 4 | 23d ago | MIT | A | 89 | Claude Code |
| developerhowar31/magi-workflow | Agent | 3 | 1mo ago | MIT | A | 83 | Claude Code |
| refactor-executorscolladon/craft | Agent | 2 | 2d ago | MIT | A | 47 | Claude Code |