Automated summary
This skill analyzes code changes after implementation, updates DESIGN.md files with learned insights, and proposes improvements.
Strengths
- Automates design document updates with accurate, real-code-derived information.
- Encourages continuous improvement by documenting lessons learned and suggesting refactorings.
- Integrates into a story-driven development workflow.
Limitations
- Requires a well-defined project structure with DESIGN.md files and feature folders.
- May produce redundant updates if changes are trivial.
- Relies on clear commit messages and diffs for accurate analysis.
Use this skill at the end of each feature or user story to capture learnings and keep documentation current.
Do not use for minor bug fixes or changes that do not affect the overall system design.
Security analysis
CautionThe skill orchestrates a development feedback loop using subagents, which involves reading and writing documentation, running git diff, and executing commands via Bash. While the primary purpose is legitimate, the use of Bash and file modifications poses a moderate risk that requires trust in the subagent instructions.
- •Uses Bash and file editing tools, which could potentially be misused if the agent instructions are compromised or not reviewed.
Examples
After finishing the user authentication feature, run feedback to update DESIGN.md with what we learned.I just completed the payment integration story. Execute dev:feedback to accumulate learnings, update the design document, and suggest any refactoring.Finished implementing the search functionality. Use feedback to analyze changes, update the design docs, and evaluate whether the TDD tests need simplification or removal.name: dev:feedback description: | 実装完了後、学んだことをDESIGN.mdに蓄積。リファクタリング/スキル・ルールの自己改善も提案。 ストーリー駆動開発の終点。 「フィードバック」「/dev:feedback」で起動。
Trigger: フィードバック, 学習事項蓄積, /dev:feedback, feedback, design update allowed-tools:
- Read
- Write
- Edit
- Bash
- Glob
- Grep
- Task
- AskUserQuestion
フィードバック → 仕様書蓄積(dev:feedback)
エージェント委譲ルール
⚠️ 分析・更新・提案は必ずTaskエージェントに委譲する。自分で実行しない。
呼び出しパターン(全ステップ共通):
agentContent = Read(".claude/skills/dev/feedback/agents/{agent}.md")
Task({ prompt: agentContent + 追加コンテキスト, subagent_type: {type}, model: {指定モデル} })
| Phase | agent | model | type | 追加コンテキスト | |-------|-------|-------|------|-----------------| | 1 | analyze-changes.md | sonnet | general-purpose | git diff, feature-slug, story-slug | | 2a | update-design.md | sonnet | general-purpose | Phase 1のJSON + feature-slug | | 2c | (code-simplifier) | opus | code-simplifier | docs/features/DESIGN.mdのパス | | 3 | propose-improvement.md | opus | general-purpose | Phase 1-2の結果 + feature-slug | | 4 | evaluate-tests.md | haiku | general-purpose | テストファイル一覧 |
出力先
| ファイル | Phase |
|----------|-------|
| docs/features/{feature-slug}/DESIGN.md | 2a |
| docs/features/DESIGN.md | 2b |
| docs/features/{feature-slug}/IMPROVEMENTS.md | 3(該当時) |
★ 実行手順(必ずこの順序で実行)
Phase 1: 変更内容の収集 → 分析JSON
- → エージェント委譲(analyze-changes.md / sonnet)
- 分析結果をJSONとして保持
ゲート: 分析JSONが得られなければ次に進まない。
Phase 2a: 機能DESIGN.md更新
- → エージェント委譲(update-design.md / sonnet)
docs/features/{feature-slug}/DESIGN.mdに追記・保存
ゲート: 機能DESIGN.mdが更新されなければ次に進まない。
Phase 2b: 総合DESIGN.md更新 ★必須★
- Phase 2aの内容から、プロジェクト全体に影響する設計判断を抽出
docs/features/DESIGN.mdに追記(Edit)- 判断基準は references/update-format.md を参照
ゲート: 総合DESIGN.mdが更新されなければ次に進まない。
Phase 2c: 総合DESIGNの整理
- → エージェント委譲(code-simplifier / opus)
- 整理観点は references/update-format.md の「総合DESIGN.md整理の観点」を参照
Phase 3: 改善提案ドキュメント化(該当時)
- → エージェント委譲(propose-improvement.md / opus)
docs/features/{feature-slug}/IMPROVEMENTS.mdに保存
Phase 4: テスト資産の整理(TDDタスクがあった場合)
- → エージェント委譲(evaluate-tests.md / haiku)— 保持/簡素化/削除を分類
- AskUserQuestion でテスト整理方針を確認(整理する/すべて保持/スキップ)
- 選択に応じてテストの簡素化・削除を実行
完了条件
- [ ] 変更内容が分析された(Phase 1)
- [ ] 機能DESIGN.mdが更新された(Phase 2a)
- [ ] 総合DESIGN.mdが更新された(Phase 2b) ★必須★
- [ ] 総合DESIGNが整理された(Phase 2c)
- [ ] 改善提案が作成された(該当時)(Phase 3)
- [ ] テスト資産が整理された(TDD時)(Phase 4)
参照
- agents/: analyze-changes.md, update-design.md, propose-improvement.md, evaluate-tests.md
- references/: design-template.md, update-format.md, improvement-patterns.md
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/feedback-design-specification-accumulation-mma3jrno/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/feedback-design-specification-accumulation-mma3jrno' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/feedback-design-specification-accumulation-mma3jrno/download' --output '.skill-downloads/feedback-design-specification-accumulation-mma3jrno/SKILL.md.part' && mv '.skill-downloads/feedback-design-specification-accumulation-mma3jrno/SKILL.md.part' '.skill-downloads/feedback-design-specification-accumulation-mma3jrno/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.
Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/feedback-design-specification-accumulation-mma3jrno/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.