Menu
从公域知识萃取方法论并封装为 Skill。通过 NotebookLM Deep Research 搜索优质内容，提炼工作流、原则、模板，自动生成标准化技能。触发词：/note-meta-skill, 从知识创建技能, 知识萃取, 元技能
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.
Based on SOC occupation classification
Craft-first interface design for dashboards, admin panels, SaaS apps, tools, settings pages, data interfaces, and interactive products. Use when designing, building, reviewing, auditing, or refining product UI where visual craft, layout hierarchy, tokens, states, visual direction, image-based references, or design-system consistency matter. Not for marketing pages, landing pages, campaigns, or brand-only work.
Tooling and conventions for JavaScript/TypeScript projects. Use when setting up new projects, configuring ESLint/Prettier alternatives, monorepos, or library publishing.
Run technical UI quality audits across accessibility, performance, theming, responsive behavior, and anti-patterns.
Comprehensive web quality audit covering performance, accessibility, SEO, and best practices. Use when asked to "audit my site", "review web quality", "run lighthouse audit", "check page quality", or "optimize my website".
Complete reference for implementing OAuth 2.0 login flows with major AI service providers: OpenAI/ChatGPT, Anthropic Claude/Claude Code, Google Gemini, and GitHub Copilot. USE THIS SKILL whenever working on authentication, login, OAuth, tokens, or API access for any of these AI services — even if the user just says "how do I log in to X API", "I need to auth with Claude/OpenAI/Gemini/Copilot", "implement OAuth for AI service", "refresh tokens", "PKCE flow", "authorization code grant", "device code flow", or anything relating to integrating user authentication into an app that uses these AI APIs. Also trigger when comparing auth patterns across AI providers, troubleshooting 401/403 errors on AI APIs, or building MCP servers/clients that need user-scoped AI tokens.
Complete reference for building apps with the Pollinations API — a multi-modal AI generation platform supporting text, image, video, audio, and embeddings. Use this skill whenever working with Pollinations, gen.pollinations.ai, pollinations.ai, or any of their models. ALWAYS trigger on: text generation with Pollinations models, image generation via URL or API, video generation, TTS/audio generation, embeddings, BYOP (Bring Your Own Pollen) OAuth flows, MCP server setup, CLI usage, account/balance/usage APIs, or safety filtering. Also trigger when the user mentions "pollen", "flux model", "pollinations image", "polli CLI", device flow for Pollinations, or wants to integrate Pollinations into any TypeScript/Node/Python/React app. If the user wants to generate any media with AI without paying OpenAI directly, this is a strong candidate — check this skill.
| name | note-meta-skill |
| description | 从公域知识萃取方法论并封装为 Skill。通过 NotebookLM Deep Research 搜索优质内容，提炼工作流、原则、模板，自动生成标准化技能。触发词：/note-meta-skill, 从知识创建技能, 知识萃取, 元技能 |
从公域互联网或用户提供的文档中萃取优质知识，提炼出工作流、方法论、原则等核心内容，自动封装为标准化的 Skill。
notebooklm 技能进行知识收集notebooklm login在所有 notebooklm 命令前设置代理：
$env:HTTP_PROXY="http://127.0.0.1:7890"
$env:HTTPS_PROXY="http://127.0.0.1:7890"
⚠️ notebooklm login 完成后，必须回到终端按 ENTER 键才能保存认证状态。
| 问题 | 解决方案 |
|---|---|
| All connection attempts failed | 设置 HTTP_PROXY/HTTPS_PROXY |
| 登录后仍报 Auth expired | 登录后必须按 ENTER |
| TargetClosedError | 设置 $env:NOTEBOOKLM_HOME="~/.notebooklm_fix" 切换新目录 |
用户输入 (主题/URL/文档)
↓
阶段1: 知识收集
↓
阶段2: 知识萃取
↓
★ 确认点 ★
↓
阶段3: Skill 封装
↓
交付 Skill
询问用户：
notebooklm create "知识萃取: [主题名称]"
方式 A - AI 自主搜索（Deep Research）：
notebooklm source add-research "[搜索查询]" --mode deep --no-wait
notebooklm research wait --import-all
方式 B - 用户提供来源：
# 添加 URL
notebooklm source add "https://..."
# 添加本地文件
notebooklm source add ./document.pdf
方式 C - 混合模式（推荐）： 先添加用户文档，再用 Deep Research 补充搜索。
notebooklm source list --json
确认所有 source 的 status = READY 后继续。
使用结构化提问从 NotebookLM 中萃取核心知识。
按顺序执行以下提问（完整 Prompt 见 extraction_prompts.md）：
| 维度 | 目的 | 核心问题 |
|---|---|---|
| Workflow | 提取步骤化流程 | "这套方法的完整工作流是什么？每一步具体做什么？" |
| Principles | 提取核心原则/方法论 | "背后的核心原则是什么？为什么要这样做？" |
| Templates | 提取可复用模板 | "有没有可以直接复用的模板、框架或清单？" |
| Scripts | 探测是否有脚本 | "有没有涉及代码、脚本或自动化工具？" |
# 依次执行萃取提问
notebooklm ask "[Workflow 萃取 Prompt]" --json
notebooklm ask "[Principles 萃取 Prompt]" --json
notebooklm ask "[Templates 萃取 Prompt]" --json
notebooklm ask "[Scripts 萃取 Prompt]" --json
将所有萃取结果整理为结构化文档，格式如下：
# 知识萃取报告: [主题名称]
## Workflow (工作流)
1. 步骤1: ...
2. 步骤2: ...
## Principles (核心原则)
- 原则1: ...（解释 Why）
- 原则2: ...
## Templates (模板)
[如有]
## Scripts (脚本)
[如有，提取代码；如无，标注"无"]
暂停并向用户展示萃取结果，确认：
根据萃取结果确定：
writing-mastermkdir -p ~/.gemini/antigravity/Skills/[skill-name]/references
使用 skill_template.md 作为基础模板，填入：
references/ 目录根据萃取内容创建：
principles.md：详细方法论templates.md：可复用模板[其他].md：按需添加如果萃取结果包含脚本：
mkdir -p ~/.gemini/antigravity/Skills/[skill-name]/scripts
将脚本代码保存到 scripts/ 目录。
最终交付物：
[skill-name]/
├── SKILL.md # 主技能文件
├── references/ # 参考资料
│ ├── principles.md # 核心原则（如有）
│ └── templates.md # 模板（如有）
└── scripts/ # 脚本（如有）
用户请求："我想学习这位写作博主的方法，帮我做成一个 Skill"
执行流程：
notebooklm create "知识萃取: 写作方法论"writing-master/SKILL.md