harness-work

HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5. Do NOT load for: planning, review, release, setup.

Jump to install

Repository: Chachamaru127/claude-code-harness
Last source activity: August 10, 2026 at 09:25
Detected SKILL.md language: Mixed languages
Stars: 3,058
Forks: 302

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/Chachamaru127/claude-code-harness --skill harness-work

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer

5 files

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

harness-release

Generic release automation for projects using Keep a Changelog + GitHub. Single confirmation gate then end-to-end automation: bump detection, CHANGELOG promotion, PR/main merge, tag, GitHub Release. Trigger: release, version bump, publish. Do NOT load for: implementation, review, planning, setup.

Generate an Acceptance Demo HTML for non-engineer vibecoders right before ship/wait/reject decision. Reads back the acceptance_criteria that were stored as personal-preference.v1 by harness-plan-brief (joined by user_request_hash), then renders a single-file HTML showing each criterion as verified or unverified along with a ship/wait/reject recommendation. Use when the user asks for an acceptance review, wants to decide whether to ship a delivered task, or says: acceptance demo, accept demo, 受け入れ判断, 受入レビュー, ship/wait/reject 判定, 検収レビュー. Do NOT load for: implementation, code review, release work.

2026-08-16

View all from this repository

Harness Work

Harness の統合実行スキル。以下の旧スキルを統合:

work — Plans.md タスクの実装（スコープ自動判断）

impl — 機能実装（タスクベース）

breezing — チームフル自動実行

parallel-workflows — 並列ワークフロー最適化

ci — CI 失敗時の復旧

Quick Reference

ユーザー入力

モード

動作

harness-work

auto

タスク数で自動判定（下記参照）

harness-work all

auto

全未完了タスクを自動モードで実行

harness-work 3

solo

タスク3だけ即実行

harness-work --parallel 5

parallel

5ワーカーで並列実行（強制）

harness-work --codex

codex

Codex CLI に委託（明示時のみ）

harness-work --breezing all

breezing

resolved backend でチーム実行（配布既定は claude、user/project default で cursor 可）

harness-work --breezing --backend cursor all

breezing

Cursor worker backend を明示してチーム実行

harness-work --breezing --backend claude all

breezing

Codex native subagent worker を明示してチーム実行

harness-work --breezing

breezing

チーム実行を強制

harness-work 3 --plan roadmap

solo

named Plans の roadmap からタスク3を実行

Execution Mode Auto Selection（フラグなし時の自動判定）

明示的なモードフラグ（--parallel, --breezing, --codex）がない場合、対象タスク数に応じて最適なモードを自動選択する:

対象タスク数

自動選択モード

理由

1 件

Solo

オーバーヘッド最小。直接実装が最速

2〜3 件

Parallel（Task tool）

Worker 分離のメリットが出始める閾値

Harness Work

Harness の統合実行スキル。以下の旧スキルを統合:

work — Plans.md タスクの実装（スコープ自動判断）

impl — 機能実装（タスクベース）

breezing — チームフル自動実行

parallel-workflows — 並列ワークフロー最適化

ci — CI 失敗時の復旧

Quick Reference

ユーザー入力

モード

動作

harness-work

auto

タスク数で自動判定（下記参照）

harness-work all

auto

全未完了タスクを自動モードで実行

harness-work 3

solo

タスク3だけ即実行

harness-work --parallel 5

parallel

5ワーカーで並列実行（強制）

harness-work --codex

codex

Codex CLI に委託（明示時のみ）

harness-work --breezing all

breezing

resolved backend でチーム実行（配布既定は claude、user/project default で cursor 可）

harness-work --breezing --backend cursor all

breezing

Cursor worker backend を明示してチーム実行

harness-work --breezing --backend claude all

breezing

Codex native subagent worker を明示してチーム実行

harness-work --breezing

breezing

チーム実行を強制

harness-work 3 --plan roadmap

solo

named Plans の roadmap からタスク3を実行

Execution Mode Auto Selection（フラグなし時の自動判定）

明示的なモードフラグ（--parallel, --breezing, --codex）がない場合、対象タスク数に応じて最適なモードを自動選択する:

対象タスク数

自動選択モード

理由

1 件

Solo

オーバーヘッド最小。直接実装が最速

2〜3 件

Parallel（Task tool）

Worker 分離のメリットが出始める閾値

resolver_backend_arg = "" if explicit_backend_value in ["claude", "codex", "cursor"]: resolver_backend_arg = "--backend {explicit_backend_value}" backend = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/resolve-impl-backend.sh\" {resolver_backend_arg}") if explicit_flag == "--cursor": backend = "cursor" if explicit_flag == "--codex": backend = "codex" if topology in ["solo", "parallel"] and backend in ["cursor", "codex"]: BASE_REF = git("rev-parse", "HEAD") WT_ID = "{task.number}-$(date +%Y%m%d-%H%M%S)-$$" worktree_path = ".claude/worktrees/{backend}-{WT_ID}" worktree_branch = "{backend}-work/{WT_ID}" bash("mkdir -p .claude/worktrees && git worktree add -b {worktree_branch} {worktree_path} {BASE_REF}") companion_prompt = "{task prompt}\n\nAfter making changes, create exactly one git commit in this worktree before returning." if backend == "cursor": companion_output = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/cursor-companion.sh\" task --write --workspace {worktree_path} \"{companion_prompt}\"") else: companion_state_file = "{worktree_path}/.claude/state/codex-primary-environment.json" companion_output = bash("HARNESS_CODEX_PRIMARY_ENV_STATE_FILE={companion_state_file} bash \"${HARNESS_PLUGIN_ROOT}/scripts/codex-companion.sh\" task --write -C {worktree_path} \"{companion_prompt}\"") latest_commit = git("-C", worktree_path, "rev-parse", "HEAD") if backend == "cursor" and git("-C", worktree_path, "status", "--porcelain") != "": git("-C", worktree_path, "add", "-A") git("-C", worktree_path, "-c", "user.name=cursor-composer", "-c", "user.email=cursor-composer@local", "commit", "--no-verify", "-m", "cursor: delegated change") latest_commit = git("-C", worktree_path, "rev-parse", "HEAD") if latest_commit == BASE_REF: raise EscalationError("{backend} companion produced no commit") worker_result = {type: "companion-result.v1", baseCommit: BASE_REF, commit: latest_commit, worktreePath: worktree_path, branch: worktree_branch, files_changed: git("-C", worktree_path, "diff", "--name-only", "{BASE_REF}..HEAD"), summary: companion_output} enter_shared_review_loop(worker_result) else: run_native_solo_or_parallel()

for task in execution_order: # B-0. backend 解決（配布 fallback は claude、user/project default で cursor 可） resolver_backend_arg = "" if explicit_backend_value in ["claude", "codex", "cursor"]: resolver_backend_arg = "--backend {explicit_backend_value}" backend = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/resolve-impl-backend.sh\" {resolver_backend_arg}") if explicit_flag == "--cursor": backend = "cursor" if explicit_flag == "--codex": backend = "codex" # B-1. sprint-contract を生成 contract_path = bash("node \"${HARNESS_PLUGIN_ROOT}/scripts/generate-sprint-contract.js\" {task.number}") contract_path = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/enrich-sprint-contract.sh\" {contract_path} --check \"DoD を reviewer 観点で確認\" --approve") bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/ensure-sprint-contract-ready.sh\" {contract_path}") # B-2. Worker 委託（worktree 分離） Plans.md: task.status = "cc:WIP" # 着手時に更新（未着手タスクは cc:TODO のまま） if backend == "cursor": BASE_REF = git("rev-parse", "HEAD") WT_ID = "{task.number}-$(date +%Y%m%d-%H%M%S)-$$" worktree_path = ".claude/worktrees/cursor-{WT_ID}" worktree_branch = "cursor-work/{WT_ID}" bash("mkdir -p .claude/worktrees && git worktree add -b {worktree_branch} {worktree_path} {BASE_REF}") print("🚀 cursor / $(bash \"${HARNESS_PLUGIN_ROOT}/scripts/model-routing.sh\" --host cursor --role worker --field model) / {branch} / {task.number}") companion_prompt = "{task prompt}\n\nAfter making changes, create exactly one git commit in this worktree before returning." companion_output = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/cursor-companion.sh\" task --write --workspace {worktree_path} \"{companion_prompt}\"") latest_commit = git("-C", worktree_path, "rev-parse", "HEAD") if git("-C", worktree_path, "status", "--porcelain") != "": git("-C", worktree_path, "add", "-A") git("-C", worktree_path, "-c", "user.name=cursor-composer", "-c", "user.email=cursor-composer@local", "commit", "--no-verify", "-m", "cursor: delegated change") latest_commit = git("-C", worktree_path, "rev-parse", "HEAD") if latest_commit == BASE_REF: raise EscalationError("cursor companion produced no commit") worker_result = {type: "companion-result.v1", baseCommit: BASE_REF, commit: latest_commit, worktreePath: worktree_path, branch: worktree_branch, files_changed: git("-C", worktree_path, "diff", "--name-only", "{BASE_REF}..HEAD"), summary: companion_output} worker_id = null elif backend == "codex": BASE_REF = git("rev-parse", "HEAD") WT_ID = "{task.number}-$(date +%Y%m%d-%H%M%S)-$$" worktree_path = ".claude/worktrees/codex-{WT_ID}" worktree_branch = "codex-work/{WT_ID}" bash("mkdir -p .claude/worktrees && git worktree add -b {worktree_branch} {worktree_path} {BASE_REF}") companion_prompt = "{task prompt}\n\nAfter making changes, create exactly one git commit in this worktree before returning." companion_state_file = "{worktree_path}/.claude/state/codex-primary-environment.json" companion_output = bash("HARNESS_CODEX_PRIMARY_ENV_STATE_FILE={companion_state_file} bash \"${HARNESS_PLUGIN_ROOT}/scripts/codex-companion.sh\" task --write -C {worktree_path} \"{companion_prompt}\"") latest_commit = git("-C", worktree_path, "rev-parse", "HEAD") if latest_commit == BASE_REF: raise EscalationError("codex companion produced no commit") worker_result = {type: "companion-result.v1", baseCommit: BASE_REF, commit: latest_commit, worktreePath: worktree_path, branch: worktree_branch, files_changed: git("-C", worktree_path, "diff", "--name-only", "{BASE_REF}..HEAD"), summary: companion_output} worker_id = null else: print("🚀 claude / native-subagent / {branch} / {task.number}") worker_id = spawn_agent({ message: "タスク: {task.内容}\nDoD: {task.DoD}\ncontract_path: {contract_path}\nspec_path: {spec_path}\nspec_skip_reason: {spec_skip_reason}\nmode: breezing\n\n作業は分離 worktree で行い、完了後に git commit してください。\n完了時は {commit, worktreePath, branch, files_changed, summary} を返してください。", fork_context: true }) worker_result = wait_agent({ targets: [worker_id] }) # worker_result には {commit, worktreePath, branch, files_changed, summary} が含まれる # backend=cursor/codex では Lead が companion stdout を companion-result.v1 に正規化する。 # B-3. Worker が advice request を返した時だけ、Lead が Advisor を呼ぶ if backend == "claude" and worker_result.type == "advisor-request.v1": advisor_id = spawn_agent({ message: worker_result.request_json, agent_type: "default", fork_context: true }) advisor_result = wait_agent({ targets: [advisor_id] }) close_agent({ target: advisor_id }) send_input({ target: worker_id, message: "advisor-response.v1: {advisor_result}" }) worker_result = wait_agent({ targets: [worker_id] }) # B-4. Lead がレビュー実行（Codex exec 優先） if backend == "claude": diff_text = git("-C", worker_result.worktreePath, "show", worker_result.commit) else: diff_text = git("-C", worker_result.worktreePath, "diff", "{worker_result.baseCommit}..HEAD") verdict = codex_exec_review(diff_text) or reviewer_agent_review(diff_text) profile = jq(contract_path, ".review.reviewer_profile") review_input = "review-output.json" if profile == "runtime": review_input = bash("cd {worker_result.worktreePath} && bash \"${HARNESS_PLUGIN_ROOT}/scripts/run-contract-review-checks.sh\" {contract_path}") runtime_verdict = jq(review_input, ".verdict") if runtime_verdict == "REQUEST_CHANGES": verdict = "REQUEST_CHANGES" elif runtime_verdict == "DOWNGRADE_TO_STATIC": pass # runtime 検証コマンドなし → static verdict をそのまま使う browser_result = "" if profile == "browser": # browser artifact から route / browser_mode / execution_instructions を再利用して browser runner を起動する。 browser_artifact = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/generate-browser-review-artifact.sh\" {contract_path}") browser_result = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/browser-review-runner.sh\" {browser_artifact}") browser_verdict = jq(browser_result, ".browser_verdict") if browser_verdict == "REQUEST_CHANGES": verdict = "REQUEST_CHANGES" elif browser_verdict == "APPROVE" and verdict != "REQUEST_CHANGES": verdict = "APPROVE" # browser_verdict == PENDING_BROWSER のときは static verdict を維持する # review_input が DOWNGRADE_TO_STATIC の場合は static review 結果を使う if review_input != "review-output.json" and jq(review_input, ".verdict") == "DOWNGRADE_TO_STATIC": review_input = "review-output.json" # static review の結果にフォールバック bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/write-review-result.sh\" {review_input} {latest_commit} --browser-result {browser_result}") # B-5. 修正ループ（REQUEST_CHANGES 時、contract の max_iterations まで） review_count = 0 # sprint-contract が存在するときのみ max_iterations を読む。存在しない場合は 3（後方互換） MAX_REVIEWS = read_contract(contract_path, ".review.max_iterations") or 3 latest_commit = worker_result.commit while verdict == "REQUEST_CHANGES" and review_count < MAX_REVIEWS: if backend == "claude": send_input({ target: worker_id, message: "指摘内容: {issues}\n修正して amend してください" }) # Worker が修正 → amend → 更新された commit hash を返す updated_result = wait_agent({ targets: [worker_id] }) latest_commit = updated_result.commit elif backend == "cursor": previous_commit = latest_commit companion_output = bash("bash \"${HARNESS_PLUGIN_ROOT}/scripts/cursor-companion.sh\" task --write --workspace {worker_result.worktreePath} \"Review findings:\n{issues}\n\nFix the findings and commit the result.\"") latest_commit = git("-C", worker_result.worktreePath, "rev-parse", "HEAD") if git("-C", worker_result.worktreePath, "status", "--porcelain") != "": git("-C", worker_result.worktreePath, "add", "-A") git("-C", worker_result.worktreePath, "-c", "user.name=cursor-composer", "-c", "user.email=cursor-composer@local", "commit", "--no-verify", "-m", "cursor: review fix") latest_commit = git("-C", worker_result.worktreePath, "rev-parse", "HEAD") if latest_commit == previous_commit: raise EscalationError("cursor companion retry produced no new commit") worker_result.commit = latest_commit worker_result.summary = companion_output else: previous_commit = latest_commit companion_state_file = "{worker_result.worktreePath}/.claude/state/codex-primary-environment.json" companion_output = bash("HARNESS_CODEX_PRIMARY_ENV_STATE_FILE={companion_state_file} bash \"${HARNESS_PLUGIN_ROOT}/scripts/codex-companion.sh\" task --write -C {worker_result.worktreePath} \"Review findings:\n{issues}\n\nFix the findings and commit the result.\"") latest_commit = git("-C", worker_result.worktreePath, "rev-parse", "HEAD") if latest_commit == previous_commit: raise EscalationError("codex companion retry produced no new commit") worker_result.commit = latest_commit worker_result.summary = companion_output if backend == "claude": diff_text = git("-C", worker_result.worktreePath, "show", latest_commit) else: diff_text = git("-C", worker_result.worktreePath, "diff", "{worker_result.baseCommit}..HEAD") verdict = codex_exec_review(diff_text) or reviewer_agent_review(diff_text) review_count++ # B-6. Worker 終了 if backend == "claude": close_agent({ target: worker_id }) # B-7. APPROVE → trunk に cherry-pick（feature ブランチ経由） # Worker の Branch Guard により trunk HEAD は動かず、commit は feature ブランチ上にある想定 if verdict == "APPROVE": TRUNK=$(git symbolic-ref refs/remotes/origin/HEAD 2>/dev/null | sed 's|refs/remotes/origin/||' || echo "main") git checkout "$TRUNK" # safety: 既に trunk なら no-op # feature ブランチの commit が既に trunk にある（Branch Guard 失敗時のフォールバック）か確認 if git("merge-base", "--is-ancestor", latest_commit, "HEAD"): pass # 既に trunk 上 — cherry-pick 不要（再入防止） else: if backend == "claude": git cherry-pick --no-commit {latest_commit} # feature branch → trunk else: git cherry-pick --no-commit {worker_result.baseCommit}..{latest_commit} # companion range → trunk git commit -m "{task.内容}" # Worker の worktree を remove してから feature ブランチを削除 if worker_result.worktreePath: git worktree remove {worker_result.worktreePath} --force if worker_result.branch and worker_result.branch not in ["main", "master"] and worker_result.branch != TRUNK: git branch -D {worker_result.branch} Plans.md: task.status = "cc:完了 [{hash}]" # auto-checkpoint 記録（冪等性ガード (c)） # Plans.md 書き換え直後に呼ぶ。失敗しても fail-open（|| true）でループを止めない HASH=$(git rev-parse --short HEAD) REVIEW_RESULT_PATH=".claude/state/review-results/${task.number}.review-result.json" bash "${HARNESS_PLUGIN_ROOT}/scripts/auto-checkpoint.sh" \ "${task.number}" "${HASH}" "${contract_path}" "${REVIEW_RESULT_PATH}" \ || true # fail-open: harness-mem 未起動環境でも継続 else: → ユーザーにエスカレーション # B-8. Progress feed print("📊 Progress: Task {completed}/{total} 完了 — {task.内容}")

name	harness-work
description	HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5. Do NOT load for: planning, review, release, setup.
description-en	HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5. Do NOT load for: planning, review, release, setup.
description-ja	HAR:Plans.md タスクを1件から全並列チーム実行まで担当。実装して、実行して、全部やって、breezing、チーム実行、parallel、composer、コンポーザー、composer 2.5 で起動。プランニング・レビュー・リリース・セットアップには使わない。
kind	workflow
purpose	Execute Plans.md tasks end to end through Codex-native tools
trigger	implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5, composer mode, コンポーザー
shape	workflow
role	executor
pair	harness-review
owner	harness-core
since	2026-05-05
allowed-tools	["Read","Write","Edit","Grep","Glob","Bash","spawn_agent","send_input","wait_agent","close_agent"]
argument-hint	[all] [task-number\|range] [--backend claude\|codex\|cursor] [--cursor] [--codex] [--parallel N] [--no-commit] [--resume id] [--breezing] [--auto-mode] [--tdd-bypass]
effort	high

name	harness-work
description	HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5. Do NOT load for: planning, review, release, setup.
description-en	HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5. Do NOT load for: planning, review, release, setup.
description-ja	HAR:Plans.md タスクを1件から全並列チーム実行まで担当。実装して、実行して、全部やって、breezing、チーム実行、parallel、composer、コンポーザー、composer 2.5 で起動。プランニング・レビュー・リリース・セットアップには使わない。
kind	workflow
purpose	Execute Plans.md tasks end to end through Codex-native tools
trigger	implement, execute, do everything, breezing, team run, parallel, composer, composer 2.5, composer mode, コンポーザー
shape	workflow
role	executor
pair	harness-review
owner	harness-core
since	2026-05-05
allowed-tools	["Read","Write","Edit","Grep","Glob","Bash","spawn_agent","send_input","wait_agent","close_agent"]
argument-hint	[all] [task-number\|range] [--backend claude\|codex\|cursor] [--cursor] [--codex] [--parallel N] [--no-commit] [--resume id] [--breezing] [--auto-mode] [--tdd-bypass]
effort	high

backend	実装の担い手	委託コマンド
`claude`（global fallback）	Codex native subagent（`spawn_agent({message, fork_context})`）	spawn_agent で worker を spawn
`codex`	Codex CLI	`bash "${HARNESS_PLUGIN_ROOT}/scripts/codex-companion.sh" task --write "<prompt>"`
`cursor`	cursor-agent（model `composer-2.5-fast`）	`bash "${HARNESS_PLUGIN_ROOT}/scripts/cursor-companion.sh" task --write --workspace <worktree> "<prompt>"`

オプション	説明	デフォルト
`all`	全未完了タスクを対象	-
`N` or `N-M`	タスク番号/範囲指定	-
`--parallel N`	並列ワーカー数	auto
`--sequential`	直列実行強制	-
`--codex`	Codex CLI で実装委託（明示時のみ、自動選択しない）	false
`--backend <claude\|codex\|cursor>`	明示バックエンド選択（worker ロールのみ適用、precedence 最上位）	resolver result（未設定時は claude）
`--cursor`	cursor backend（`--backend cursor` の別名）	false
`--plan NAME`	`plans/manifest.json` の named plan を使う	active/default
`--no-commit`	自動コミット抑制	false
`--resume <id\|latest>`	前回セッション再開	-
`--breezing`	Lead/Worker/Reviewer のチーム実行	false
`--no-tdd`	TDD フェーズスキップ	false
`--tdd-bypass`	緊急時だけ TDD 強制を bypass。`HARNESS_TDD_BYPASS_REASON` または明示理由を audit に残す	false
`--no-simplify`	Auto-Refinement スキップ	false
`--auto-mode`	Auto Mode rollout を明示。親セッションの permission mode が互換な場合のみ採用を検討	false

詳細	参照
Codex native Solo / Parallel / Breezing の具体手順	`references/execution-modes.md`
companion review、Reviewer fallback、AI Residuals、修正ループ	`references/review-loop.md`
完了報告の生成	`references/completion-report.md`
テスト/CI 失敗時の再チケット化	`references/failure-reticketing.md`
仕様正本チェックの基準	`docs/plans/spec-ssot.md`

要素	条件	スコア
ファイル数	変更対象 4 ファイル以上	+1
ディレクトリ	core/, guardrails/, security/ を含む	+1
キーワード	architecture, security, design, migration を含む	+1
失敗履歴	agent memory に同タスクの失敗記録あり	+2
明示指定	PM テンプレートに `effort: high` / `effort: xhigh`（旧 `ultrathink` も互換受理）記載あり	+3（自動採用）

スコア	code-risk（core/guardrails/security/architecture/migration を含む）	effort tier
0-2	不問	`medium`（Worker frontmatter 既定のまま）
≥ 3	なし	`high`
≥ 3	あり	`xhigh`

条件	アクション
`cc:完了` 後にテスト失敗	修正タスク案を state に保存し、承認を待つ
CI 失敗（3回未満）	修正を実施し、失敗カウントをインクリメント
CI 失敗（3回目）	修正タスク案を提示 + エスカレーション

重要度	定義	verdict への影響
critical	セキュリティ脆弱性、データ損失リスク、本番障害の可能性	1 件でも → REQUEST_CHANGES
major	既存機能の破壊、仕様との明確な矛盾、テスト不通過	1 件でも → REQUEST_CHANGES
minor	命名改善、コメント不足、スタイル不統一	verdict に影響しない
recommendation	ベストプラクティス提案、将来の改善案	verdict に影響しない

公式 plugin	Harness	verdict 影響
`approve`	`APPROVE`	-
`needs-attention`	`REQUEST_CHANGES`	-
`findings[].severity: critical`	`critical_issues[]`	1件でも → REQUEST_CHANGES
`findings[].severity: high`	`major_issues[]`	1件でも → REQUEST_CHANGES
`findings[].severity: medium/low`	`recommendations[]`	verdict に影響しない

harness-work

Source facts

Install options

Review the source files

More from this repository

More from this repository

Harness Work

Quick Reference

Execution Mode Auto Selection（フラグなし時の自動判定）

Harness Work

Quick Reference

Execution Mode Auto Selection（フラグなし時の自動判定）

ルール

Execution Backend Selection（実装バックエンド選択）

解決手順

自然言語 backend trigger

role-scoped 制約

非 claude バックエンドの self_review ゲート

cursor バックエンドの banner（委託前に必須）

オプション

Progressive Disclosure

重要停止条件

スコープダイアログ（引数なし時）

Effort レベル制御（Opus 4.8 / v2.1.111+）

多要素スコアリング

effort tier の決め方（注入しない）

実行モード詳細

Harness helper script root

Backend-resolved executor path (Solo / Parallel / Breezing)

Solo モード（1 件時の自動選択）

Parallel モード（2〜3 件時の自動選択 / --parallel N で強制）

Codex モード（--codex 明示時のみ）

Breezing モード（4 件以上で自動選択 / --breezing で強制）

Work Mode Lifecycle (bin/harness work-mode)

Advisor Protocol（全モード共通）

Solo モードでの Advisor

Sprint Contract

CI 失敗時の対応

失敗タスクの自動再チケット化

トリガー条件

修正タスクの自動生成

レビューループ

レビュー実行の優先順位

APPROVE / REQUEST_CHANGES の判定基準

Codex exec レビュー（公式プラグイン経由）

内部 Reviewer agent フォールバック

修正ループ（REQUEST_CHANGES 時）

Breezing モードでの適用

Completion Report Output Contract

関連スキル

非 `claude` バックエンドの self_review ゲート

Parallel モード（2〜3 件時の自動選択 / `--parallel N` で強制）

Codex モード（`--codex` 明示時のみ）

Breezing モード（4 件以上で自動選択 / `--breezing` で強制）

Work Mode Lifecycle (`bin/harness work-mode`)