skills/continuous-learning/SKILL.md
Extracts reusable patterns from Claude Code sessions and captures knowledge as atomic memory units, then evaluates whether each memory should be upgraded to framework-shared rules, methodologies, or error-patterns. Use when session ends (Stop hook), when recording technical decisions, implementation insights, or lessons learned. Handles automatic pattern detection, structured memory capture with interconnected knowledge links, and post-write upgrade decision flow to prevent cross-project principles being trapped in single-project memory.
npx skillsauth add tarrragon/claude continuous-learningInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
從 Claude Code 工作過程中自動提取可復用模式,並將洞察、決策和經驗記錄為結構化的原子記憶單位。
透過 Stop hook 在 session 結束時自動執行:
.claude/skills/learned/將重要技術決策、實作方案和經驗教訓記錄為原子化記憶:
重要:memory 寫入不是終點,而是升級評估的起點。寫入
feedback_*.md後必須執行升級評估,否則跨專案通用原則會被困在單一專案的 memory 中(PC-061)。
- 升級評估規則:
.claude/pm-rules/pm-quality-baseline.md規則 7- 錯誤模式參考:
.claude/error-patterns/process-compliance/PC-061-memory-upgrade-blindness.md- 完整決策樹:
references/upgrade-decision-tree.md
適用時機:
| 時機 | 說明 | |------|------| | 重要技術決策完成 | 方案選擇後建立決策記錄 | | 實作方案確定 | 新的實作模式或解決方案誕生 | | 學習機會 | 測試失敗、問題排除、重構完成後的經驗總結 | | Phase 4 完成 | 重構後進行知識沉澱 | | 版本發布前 | 總結主要決策和經驗 |
當記錄的 memory 核心是根因分析(error-pattern、代理人失敗歸因、用戶質疑「分析太表層」),必須套用兩階段深度反思:
禁止只列 1-2 個假設就下結論,或跳過 Phase 2 直接落地。
完整方法論:
.claude/methodologies/three-phase-reflection-methodology.md案例:PC-087(表層版)→ PC-088(Phase 1+2 後的抽象層)
| Pattern | Description |
|---------|-------------|
| error_resolution | How specific errors were resolved |
| user_corrections | Patterns from user corrections |
| workarounds | Solutions to framework/library quirks |
| debugging_techniques | Effective debugging approaches |
| project_specific | Project-specific conventions |
Edit config.json to customize:
{
"min_session_length": 10,
"extraction_threshold": "medium",
"auto_approve": false,
"learned_skills_path": ".claude/skills/learned/",
"patterns_to_detect": [
"error_resolution",
"user_corrections",
"workarounds",
"debugging_techniques",
"project_specific"
],
"ignore_patterns": ["simple_typos", "one_time_fixes", "external_api_issues"]
}
Add to .claude/settings.json:
{
"hooks": {
"Stop": [
{
"matcher": "*",
"hooks": [
{
"type": "command",
"command": "$CLAUDE_PROJECT_DIR/.claude/skills/continuous-learning/evaluate-session.py"
}
]
}
]
}
}
記憶建立的完整規範(類型定義、結論式標題設計、標準結構、連結策略、原子性原則):
參考: references/memory-capture-guide.md
完成 Step 4「儲存到 memory/」後,禁止直接結束。必須對寫入的 memory 執行升級評估:
| 步驟 | 動作 | 工具/參考 |
|------|------|----------|
| 5.1 | 對每個新建的 feedback_*.md 檔案執行四問檢查 | .claude/pm-rules/pm-quality-baseline.md 規則 7 |
| 5.2 | 判斷升級目的地(六類分支) | references/upgrade-decision-tree.md |
| 5.3 | 執行升級寫入(rules / pm-rules / error-patterns / methodologies / references / skills) | 對應目錄 |
| 5.4 | 在原 memory 檔案頂部加註「已升級」標註 | 標註格式見決策樹 |
為什麼必須執行:
memory 寫入不是終點,而是升級評估的起點。若略過此步驟,跨專案通用的原則會被困在單一專案的 auto-memory 中,無法 sync 到其他專案,也不會被其他 session 自動載入(PC-061「Memory upgrade blindness」)。
參考資源:
.claude/pm-rules/pm-quality-baseline.md 規則 7「Memory 寫入必須評估跨專案升級」.claude/error-patterns/process-compliance/PC-061-memory-upgrade-blindness.mdreferences/upgrade-decision-tree.md/learn command - Manual pattern extraction mid-sessionLast Updated: 2026-04-13 Version: 2.1.0 - 新增 Step 5 升級評估,將 memory 寫入串接到 framework 升級流程(防範 PC-061)
development
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
development
Claude Code release notes 框架影響評估工具。比對 last-reviewed 版本篩出新版本,逐項分類(對框架有幫助 / 需評估 / 無影響 / 不適用),對採用項引導建 ANA + WRAP + spawn 落地。Use when: 執行 /release-notes 看到新版本、定期檢查 CC 更新、評估新功能對專案框架的影響時。Triggers: release notes, release-notes, CC 更新, claude code 更新, 版本更新評估, 新功能評估, 框架影響評估。
development
Assertion design judgment framework for flaky and design-quality issues. Use when writing tests, reviewing assertions, diagnosing flaky tests, or deciding if a timing/float/cache assertion is appropriate. Do NOT use for API syntax or refactoring.
tools
Chrome Extension 實機測試與 debug 工作流,以 chrome-devtools-mcp 為核心工具。Use when: (1) 完成功能後實機驗證 / manual test / 試看看 / 跑看看 / verify feature, (2) extension debug / popup 不作動 / content script 不注入 / service worker 報錯 / background 出問題, (3) 安裝 unpacked extension / load unpacked / 載入未封裝, (4) 看 console / 看 network / 看 log / view console / inspect requests, (5) 功能更新後重新載入 extension / rebuild reload / reload extension。涵蓋 Manifest V3 service worker / content script / popup / options page 的 chrome-devtools-mcp 工具呼叫流程。不取代 Puppeteer / Playwright 自動化 E2E(CI 用),定位為開發期手動驗證與 LLM-assisted debug。