Opportunity Factory

ニーズ発見から小さな成果物、レビュー、計測、学習までを回す汎用ワークフロー。

When to Use

• 「ニーズを探して、レビューして、作る」を継続的に回したい
• スマホアプリ、Steam ゲーム、SaaS、教材、社内改善などを量産したい
• アイデアをラバーダックで壁打ちし、実行可能な queue に分解したい
• 単発回答ではなく、反復可能な workspace factory を設計・運用したい
• テーマだけを与えて、workspace、状態管理、定期ワークフロー、レビューゲートを自律設計させたい
• 2〜3 個の定期プロンプトで状態を見ながら継続改善したい
• /Refine-Product-100 all のように多数の対象を複数passで改善したい

When Not to Use

• 1 回だけの実装、調査、レビューなら通常対応でよい
• 常時守るコーディング規約なら instruction にする
• 特定 persona や tool 制限が主目的なら custom agent にする
• deterministic な同期、変換、検証だけなら script / hook にする

Core Idea

抽象ループは次の通り。

discover -> research -> evaluate -> design -> build -> review -> launch/track -> learn

各ループは「次に作るもの」ではなく「どの痛み・需要を検証するか」から始める。

Workflow

1. Frame the factory 対象ドメイン、成果物タイプ、成功指標、禁止事項を 5 行以内で定義する。
2. Rubber-duck the intent 前提、誰の痛みか、既存代替、最小検証、失敗条件を質問で露出させる。
3. Design the workspace loop ユーザーがテーマだけを与えた場合は、状態ファイル、dashboard、Top-N/portfolio、候補深掘り、GO 後の product maturation、週次 workflow review、health reconciler まで含む自走ループを設計する。
4. Create the queue 作業を discover|research|evaluate|design|build|review|track|learn に分ける。
5. Produce artifacts 各 task は 1 つの evidence artifact を残す。判断、根拠、次アクションを分離する。
6. Run review gates UX、技術、法務/規約、配布、収益/成果指標をドメインに合わせて確認する。
7. Track outcomes 実測、推定、未確認を区別し、当たりだけを次サイクルで厚くする。

Operating Rules

• 先に「誰のどんな未充足ニーズか」を固定してから解決策を作る。
• 1 task は 1 artifact に収まる粒度にする。
• Advisory-only schedules are safe but too slow for a real factory; if the user expects progress, add at least one bounded mutating worker that writes one artifact and updates local state.
• 複数 worker で回す場合は、discovery/research worker、build/decision worker、reporter-learner など役割を分け、各 run は 1 task / 1 artifact / 明示的 state 更新に制限する。
• 小さい queue では commander/worker を別々にスケジュールせず、1本の single-cycle automation で回してよい (references/runtime-modes.md ## Single-Cycle Automation)。
• Mutating workers need atomic duplicate-run prevention (create-new/O_EXCL or a transactional lease, never check-then-create), a worker-wide singleton lock in addition to per-task locks, and a partial-run recovery path that reconciles artifacts instead of trusting a stale in_progress label; see references/lifecycle-and-health.md ## Lock Scopes.
• Run the differentiation gate before full evidence collection and build, not as part of the final review; a light probe is allowed, but a candidate that cannot beat existing substitutes must be rejected or re-angled before it consumes the expensive lanes.
• Gates must recompute from raw artifact data, not parse the producer's own summary; require machine-comparable types, externalize threshold-setting labels to durable state, and detect near-duplicate items.
• For user-facing artifacts, carry design and actual-render acceptance through the producing lanes, not only the workflow reviewer; generated screenshots or passing builds are not visual acceptance (references/workflow.md -> Rendered-UI Acceptance).
• A blocking or required-fix review starts a durable repair -> validate -> independent re-review loop; repairs never self-certify, consume the existing persistence budget, and retain finding-level evidence (references/rubber-duck-review.md).
• Maintain a canonical dashboard/status state for future sessions and user status answers; every workflow that changes artifacts, queues, gates, portfolio ranking, blockers, or schedules must update it with backup + stale-write checks.
• Add a workflow-review loop as a first-class workflow for self-improving factories; it reviews cadence, queue quality, Top-N replacements, dashboard drift, missing gates, and unsafe autonomy at a slower cadence than workers.
• Portfolio factories need a Top-N state with explicit replacement rules; do not grow candidate lists forever, and do not replace an incumbent without comparative evidence and reviewer critique.
• Separate reviewer, queue-eligibility, and human-approval gates: once the user approves an autonomy envelope, local/private work inside it may advance on reviewer PASS without repeatedly asking the user.
• When a selected task hits an approval boundary, record it as blocked with the exact approval needed, then run or create one safe fallback task instead of stalling the factory.
• Do not rely on chat history for factory state; write a README/resume contract that tells future sessions which state, queue, outcome, log, and artifact files to read first.
• Keep prototype/build lanes in the same factory when the user wants end-to-end production, but gate source generation on a structured candidate-level continue decision plus an MVP boundary artifact.
• End-to-end factories need an explicit post-GO maturation lane to private release-readiness; do not let automation stop at graybox GO or silently cross into public release.
• Add a bounded health reconciler for JSON/dashboard/prompt/schedule/lock drift; it may auto-repair reversible local drift but must not change criteria, decisions, approval boundaries, or cadence.
• If workflows may edit their own prompts or state, enforce the improvement invariants in a guard outside the editable file set, with hash-bound candidates, a critic receipt, change-rate caps, an audit log, and a halt a human must clear; a rule written only inside an editable file is removed by one apply (references/prompt-self-improvement.md).
• If the current host cannot build-verify the target platform, mark generated code as compile-oriented and verification-blocked; never report it as built, running, or tested without a platform verification artifact.
• 指標は実測、推定、仮説を明示して混ぜない。
• 課金、ログイン、外部公開、個人情報、法的リスクは人間承認の境界にする。
• ワークスペース固有パスや秘密情報を skill 本体へ埋め込まない。

AI-Autonomous Operation

AUTO 既定、承認は security-approve のみ。hard rule と reference default の区別は §Tunable vs Hard Rules、詳細は各 references。

#	章	骨子	詳細
A	Approval Policy	2 バケット (auto / security-approve)。金銭発生は後者の 1 例。AI usage は skill 対象外。Backup-First で reversible は auto。	references/approval-policy.md
B	Autonomy Mode	Normal / AUTO 既定 / FULL / ALL の 4 段階。setup Phase 0 で mode 未指定なら AUTO 提案 + 確認。secret 露出等は全 mode で対象外。	references/runtime-modes.md (ai-autonomous preset)
C	Fallback Lane	blocked/stall/idle 時 10 lane 順次 auto-dequeue (1 Portfolio → 2 Prompt review → 3 Advisory Critic → 4 Anti-pattern → 5 Discovery → 6 Small-Bet → 7 Learning → 8 Cleanup → 9 Real-surface RO → 10 Digest)。Discovery Floor 5 サイクル。browser 書込みは defer。	references/fallback-lane.md
D	Genuine Blocker Test	failed/stall で即 blocker 認定せず 4 問 gate (外部 signal 確認 / 別 approach N / replan / 制御不能)。4/4 Yes のみ HITL、以外は fallback へ。	references/fallback-lane.md
E	Persistence Profile	Standard / Persistent (既定) / Exhaustive。task class 別マッピング。cost/quota は skill 対象外 (adapter 任せ)。worker は自分で approach 増やさず commander が replan。	references/persistence-profile.md
F	Cadence + Adapter	worker=hourly / workflow-review=weekly + ad-hoc trigger / digest=daily。per-hour override 可。Adapter は環境依存 (Copilot Scheduler / Scout / OpenClaw / Copilot App / GH Actions / Task Scheduler / cron)。Push cadence は setup で 1 度質問、既定 manual。	references/runtime-modes.md
G	Goal + Focus Theme	無限稼働、停止は user 明示のみ。Setup で north-star + focus theme (3 ヶ月、workspace override 可) の 2 段合意。Theme apply は Layer 3 blocking critic gate (hard rule)。Candidate 完了 = Top-N 自然消滅 + shipped 明示。	references/workspace-setup.md + references/rubber-duck-review.md

Tunable vs Hard Rules

Skill は hard rule (変更不可) と reference default (AI/workspace が実運用で改善可) の 2 層。Hard rules: approval bucket 構造 (auto/security-approve)、backup-first 原則、blocker test 4 問 gate、critic 3 layer、layer 3 blocking gate 対象 5 種、substantive repair 後の独立 re-review、fallback lane auto-refill、north-star + focus theme 合意事実、independence 契約。詳細と reference default 一覧: references/tunable-defaults.md

Output Modes

Rubber Duck / Factory Plan / Workspace Setup / Self-Designing Workspace Setup / Run Slice / Periodic Runtime / Throughput Setup / Prototype Lane / Review — 用途別に責務を切替。Prototype Lane は validated candidates only + dummy data + WIP 制限 + platform-verification 必須。

References

• AI-Autonomous 基盤: rubber-duck-review.md (Layer 1/2/3 + Layer 3 SSOT) / approval-policy.md / fallback-lane.md / persistence-profile.md / tunable-defaults.md / runtime-modes.md (ai-autonomous preset)
• Workflow / Setup: workflow.md / workspace-setup.md / self-designing-factory.md / lifecycle-and-health.md / battle-tested-patterns.md / prompt-self-improvement.md / batch-refinement.md / sqlite-state-store.md
• State: dashboard-state.md / assets/templates/dashboard-state.json / assets/templates/factory-state.json / assets/templates/factory-state.sqlite.sql / assets/templates/first-run-queue.json / assets/templates/task.json / assets/templates/artifact.md
• Prompts: assets/prompts/commander.md / assets/prompts/worker.md / assets/prompts/reporter-learner.md
• Scripts: scripts/validate_factory_skill.py / scripts/init_factory_workspace.py / scripts/init_factory_sqlite.py / scripts/smoke_test_initializers.py
• Templates / Examples: assets/templates/factory-plan.md / assets/templates/setup-preflight.md / assets/examples/setup-packets.md

Done Criteria

• Primitive choice is still Skill, not prompt/instruction/agent/hook
• Domain, artifact type, success metric, and constraints are explicit
• Target surfaces, state store, prompt runner, schedule capability, and approval boundaries are explicit
• Queue has at least one next executable task
• Every task has an artifact contract and review gate
• Human approval boundaries are named
• Reviewer PASS can advance local/private work inside the approved autonomy envelope; human approval is reserved for boundaries explicitly listed in policy
• A canonical status dashboard or equivalent resume contract exists, and status answers use it first
• Self-improving factories include a scheduled or manual workflow-review loop
• Prompt self-improvements are workspace-local, evidenced by an artifact, reversible, and dashboard-recorded
• Self-editing factories enforce those invariants outside the editable file set, and a scheduled job is registered disabled, verified, then enabled with durable state already true
• Portfolio factories define Top-N capacity, replacement criteria, and demotion/watchlist/rejected states
• End-to-end factories define promotion and post-GO maturation stages through private release-readiness, with WIP/slice/retry caps and independent review
• A health reconciler detects state/prompt/schedule drift and repairs only reversible local inconsistencies
• Interrupted mutating runs can recover stale locks and partial outputs without self-certifying missing verification evidence
• If scheduled progress is expected, at least one safe mutating worker exists; advisory-only automation is called out as intentionally slow
• Approval-boundary blockers do not empty the run: the runtime records the blocker and keeps at least one safe fallback lane available
• Future sessions can resume from durable state files without reading the original chat transcript
• Prototype/source-generation tasks require candidate-level continue state, an MVP boundary artifact, dummy data only, WIP limits, and honest platform-verification status