USP
Unlike manual n8n documentation searches, this skill provides AI assistants with a multi-dimensional priority ranking system and node compatibility analysis, enabling intelligent recommendations and faster workflow creation across Claude p…
Use cases
- 01AI-assisted n8n workflow design
- 02Querying n8n node features and parameters
- 03Exploring node combinations for specific tasks
- 04Learning n8n workflow best practices
- 05Integrating AI with workflow automation
Detected files (1)
CLAUDE.mdclaude_mdShow content (9223 bytes)
# CLAUDE.md This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository. ## 專案概述 這是一個 n8n Skill Pack 生成器,將 n8n 工作流程節點資訊轉換為 AI 助理可用的 Skill Pack 格式。專案從 n8n NPM 套件收集節點資訊、從 n8n.io API 收集使用統計、從 n8n-docs 收集文件,然後組織、排序並生成結構化的 Skill Pack。 ## 常用開發指令 基本開發流程: - `npm run build` - 編譯 TypeScript 到 dist/ - `npm run build:full` 或 `npm start` - 完整建置流程,執行收集、解析、組織和生成 - `npm run dev` - TypeScript watch 模式 - `npm run clean` - 清除 dist、output 和 data/cache 目錄 - `npm run rebuild` - 清除後重新建置 測試與驗證: - `npm test` - 執行 Jest 測試 - `npm run test:watch` - Jest watch 模式 - `npm run test:coverage` - 測試覆蓋率報告 - `npm run lint` - ESLint 檢查 - `npm run typecheck` - TypeScript 型別檢查 - `npm run validate` - 驗證輸出檔案(執行 dist/scripts/validate-output.js) 更新 n8n 資料: - `npm run update` - 更新 n8n 節點資料 - `npm run update:check` - 檢查更新但不實際執行(dry-run) - `npm run update:community` - 更新社群節點資料(從 npm 搜尋熱門社群套件並解析節點詳情) - `npm run update:community:check` - 檢查社群節點更新但不實際執行 - `npm run update:website` - 更新網站資料(節點數量、n8n 版本、更新日期等) ## 核心架構 專案採用模組化架構,分為七個主要層級: ### 1. Collectors(收集器)- src/collectors/ 負責從不同來源收集原始資料: - NpmCollector: 從 n8n NPM 套件載入節點類別和基本資訊 - ApiCollector: 從 n8n.io API 收集範本和使用統計 - CommunityCollector: 從 npm registry 搜尋熱門社群節點套件,並動態安裝解析節點詳情 ### 2. Parsers(解析器)- src/parsers/ 解析和提取結構化資訊: - NodeParser: 解析節點類別的 description 物件,提取 displayName、description、version 等 - PropertyParser: 解析節點的 properties 陣列,提取輸入欄位、operations、resources 等配置 - InputOutputParser: 解析節點的輸入輸出配置,建立相容性矩陣用於判斷節點間連接規則 - DocsParser: 解析 Markdown 文件,提取摘要、關鍵字、標籤和範例 ### 3. Organizers(組織器)- src/organizers/ 組織和排序節點: - PriorityRanker: 使用多維度評分系統(使用次數、文件完整度、類別重要性)計算節點優先級 - CategoryOrganizer: 將節點分配到預定義類別(input、output、transform、trigger 等) - NodeGrouper: 依據使用頻率和功能關係將節點分組 ### 4. Generators(生成器)- src/generators/ 生成最終輸出檔案: - SkillGenerator: 生成主要的 SKILL.md 檔案,包含前 N 個最重要的節點,以及 guides/ 目錄下的指南檔案 - ResourceGenerator: 為每個節點生成獨立的詳細文件到 resources/ 目錄,採用分層合併策略 - TemplateGenerator: 從收集的範本生成 templates/ 文件 - ConnectionRuleGenerator: 生成節點相容性矩陣文件,說明節點間的連接規則 - CommunityGenerator: 為社群節點生成 community/ 目錄下的文件 ### 5. Analyzers(分析器)- src/analyzers/ 分析節點和工作流程資料: - CompatibilityAnalyzer: 分析節點間的輸入輸出相容性,建立相容性矩陣 - WorkflowAnalyzer: 分析工作流程結構和節點使用模式 ### 6. Validators(驗證器)- src/validators/ 驗證生成的輸出: - CompletenessChecker: 檢查輸出檔案的完整性 - SkillValidator: 驗證 Skill Pack 格式正確性 ### 7. Scripts(建置腳本)- scripts/ 編排完整的建置流程: - build.ts: 主要建置編排腳本,依序執行收集、解析、組織和生成步驟,使用快取機制加速重複建置 - update-n8n-data.ts: 更新 n8n 節點資料的腳本 - update-community.ts: 更新社群節點資料,從 npm 搜尋熱門套件並解析節點詳情存入快取 - update-website.ts: 更新專案網站的統計資料(節點數量、n8n 版本、更新時間等) - validate-output.ts: 驗證輸出檔案完整性和正確性 測試與除錯腳本(scripts/ 目錄下的 test-*.ts): - test-npm-collector.ts: 測試 NPM 套件收集功能 - test-io-parser.ts: 測試輸入輸出解析器 - test-compatibility.ts: 測試相容性分析器 - test-workflow-analyzer.ts: 測試工作流程分析功能 ## 建置流程 執行 `npm run build:full` 時的完整流程: 1. 收集節點資訊(npm-collector):從 n8n-nodes-base 等套件載入所有節點 2. 收集節點詳細屬性(property-parser):解析每個節點的 properties 配置 3. 收集使用統計(api-collector):從 n8n.io API 取得範本和使用次數 4. 組織和排序(priority-ranker):計算優先級分數,選出前 N 個主要節點 5. 建立相容性矩陣(compatibility-analyzer):分析節點間的輸入輸出相容性 6. 生成資源檔案(resource-generator):使用分層合併策略,高優先級節點生成獨立檔案,低優先級節點合併為分類檔案 7. 生成相容性矩陣文件(connection-rule-generator):生成節點連接規則說明 8. 獲取並生成範本檔案(template-generator):從 n8n.io API 獲取前 20 個熱門範本並生成文件 9. 生成社群節點文件(community-generator):讀取快取生成社群節點文件 10. 生成主 Skill 文件和指南(skill-generator):生成 SKILL.md 和 guides/ 目錄下的指南檔案 快取機制: - 建置過程會將中間結果儲存到 data/cache/,加速後續建置 - 快取檔案包括:nodes.json、properties.json、usage-stats.json、templates.json、node-io-config.json、compatibility-matrix.json、community-nodes.json - 使用 `npm run clean` 清除快取(但會保留 community-nodes.json,因為社群節點解析耗時較長) - CI 環境會自動偵測並啟用記憶體優化模式,定期觸發垃圾回收 ## 設定檔 - config/skill-config.json: Skill Pack 主要配置(版本、節點數量限制、分類定義、分層合併策略設定) - config/priorities.json: 優先級評分權重和類別重要性設定 - config/categories.json: 節點類別定義和分類規則 - config/community-packages.json: 社群節點套件清單(由 update:community 自動生成) ## 輸出結構 執行建置後會在 output/ 目錄生成: ``` output/ ├── SKILL.md # 主要技能檔案(包含前 10 個最重要節點) └── resources/ # 詳細節點文件 ├── INDEX.md # 所有節點的完整索引 ├── compatibility-matrix.md # 節點相容性矩陣(說明節點間連接規則) ├── guides/ # 使用指南 │ ├── how-to-find-nodes.md # 如何查找節點 │ └── usage-guide.md # 使用指南 ├── community/ # 社群節點文件 │ ├── README.md # 社群節點索引 │ └── [package-name].md # 各社群套件詳細文件 ├── input/ # 輸入類節點 ├── output/ # 輸出類節點 ├── transform/ # 轉換類節點 ├── trigger/ # 觸發類節點 ├── organization/ # 組織類節點 ├── misc/ # 其他節點 └── templates/ # 範本文件(前 20 個熱門範本) ``` ## 重要技術細節 TypeScript 編譯: - tsconfig.json 配置為編譯到 dist/,保持原始目錄結構 - 使用 Node.js >= 18.0.0 - 依賴 n8n 核心套件來存取節點型別定義 資料收集的挑戰: - 節點載入使用動態 require,需處理不同套件的載入方式 - 部分節點可能缺少文件或統計資料,需要容錯處理 - API 收集會從 n8n.io 獲取範本資料,每次請求間隔 0.5 秒以避免過載 - 社群節點收集採用動態安裝策略:安裝 → 解析 → 卸載,避免依賴衝突 相容性矩陣功能: - 解析每個節點的輸入輸出類型(main、ai_agent、ai_tool、ai_document 等) - 建立節點間的連接相容性規則,用於工作流程設計建議 - 支援多輸入/多輸出節點和動態輸出節點的分析 CI/CD 最佳化: - 自動偵測 CI 環境(CI=true 或 GITHUB_ACTIONS=true) - 在 CI 環境中啟用記憶體優化模式,定期執行垃圾回收 - 建置失敗時會嘗試從快取恢復,提高容錯性 生成的 Skill Pack: - 主 SKILL.md 包含前 10 個(可設定於 config/skill-config.json)最重要節點的簡要資訊 - resources/ 下的詳細文件按類別組織,採用分層策略:高優先級節點獨立檔案,低優先級節點合併為分類檔案 - 包含節點相容性矩陣和前 20 個熱門工作流程範本 - 包含前 30 個熱門社群節點套件的詳細文件 - 包含使用指南(how-to-find-nodes.md、usage-guide.md) - 支援 Claude Code、Claude.ai Web 和 Claude Desktop
README
n8n Skills
Supported n8n version: v2.15.1
n8n Skills is an automation skill pack that enables AI assistants to directly understand and operate n8n workflows, covering comprehensive information on 545 nodes and 20 curated templates. Through n8n Skills, AI assistants can help you quickly design workflows and explore node functionalities, making workflow automation more intuitive, intelligent, and saving significant development time.
Why Choose n8n Skills
n8n Skills provides a complete n8n Skill Pack solution for AI assistant workflow integration:
- Covers detailed documentation and usage guides for 545 n8n nodes
- Provides 20 curated workflow templates covering common scenarios like AI chatbots, data processing, and communications
- Multi-dimensional priority ranking system ensuring the most useful information is presented first
- Node compatibility analysis to help design correct workflow connections
- Supports Claude Code, Claude.ai Web, and Claude Desktop platforms
- Continuously maintained and updated to follow the latest n8n versions
With n8n Skills, you can:
- Quickly query any node's features, parameters, and configuration methods
- Have AI assistants recommend suitable node combinations for specific tasks
- Learn workflow design best practices and common patterns
- Explore unlimited application possibilities with 545 nodes
Who Should Use
n8n Skills is designed for the following users:
AI Assistant Developers
- Engineers developing with Claude Code CLI tool
- Users seeking n8n knowledge in Claude.ai web version
- Claude Desktop application enthusiasts
n8n Workflow Users
- Automation engineers who want AI-assisted workflow design
- n8n developers who need quick node functionality queries
- Learners hoping to explore node best practices
Automation Enthusiasts
- Innovators looking to integrate AI with workflow automation
- Efficiency seekers aiming to boost productivity
Project Overview
This package provides AI assistants with the ability to access n8n node information, helping AI understand and operate n8n workflows.
Key Features
- Provides complete information on n8n nodes
- Supports node search and exploration
- Node configuration validation
- Workflow structure analysis
- Support for 30+ popular community packages
- Community packages organized by category (AI tools, communication, web scraping, etc.)
Technical Architecture
This project is built upon the n8n-mcp architecture, converted into an n8n Skill Pack generator with added priority ranking, node grouping, and documentation integration features. n8n Skills adopts a five-layer modular architecture (collectors, parsers, organizers, generators, build scripts), automatically collecting information from n8n NPM packages, APIs, and documentation, and generating skill packs best suited for AI assistants through intelligent ranking.
Using n8n Skills
This project generates the n8n Skills automation skill pack, allowing you to use comprehensive n8n workflow knowledge in Claude Code, Claude.ai, or Claude Desktop.
Download n8n Skill Pack
- Go to the GitHub Releases page of this project
- Download the latest version of the
n8n-skills-{version}.zipfile - After extraction, you will get the following file structure:
n8n-skills/ ├── SKILL.md # Main skill file └── resources/ # Detailed node documentation ├── input/ # Input category nodes ├── output/ # Output category nodes ├── transform/ # Transform category nodes ├── trigger/ # Trigger category nodes ├── organization/ # Organization category nodes ├── misc/ # Miscellaneous nodes ├── community/ # Community package nodes └── templates/ # Workflow templates
Installation Methods
Choose the n8n Skills installation method according to the Claude platform you are using:
Claude Code (CLI Tool)
Suitable for developers using Claude Code in the terminal, loading n8n Skills through local skills directory.
-
Create a
.claude/skills/directory in your project root:mkdir -p .claude/skills/n8n-skills -
Copy the extracted
SKILL.mdandresources/directory to that directory:cp -r n8n-skills/* .claude/skills/n8n-skills/ -
The directory structure should look like this:
your-project/ └── .claude/ └── skills/ └── n8n-skills/ ├── SKILL.md └── resources/ -
Verify installation: Ask Claude Code "List available n8n nodes". If the Skill is correctly invoked, the installation is successful.
Claude.ai Web (Web Version)
Suitable for general users in browsers.
- Log in to Claude.ai
- Go to the "Settings" page and find the "Capabilities" section
- Click "Upload skill"
- Select the downloaded
n8n-skills-{version}.zipfile to upload - After upload completes, you will see "n8n-skills" below. If not enabled, click to enable it.
- Return to the conversation window and ask questions about n8n. Successfully using n8n-skills indicates successful installation.
Claude Desktop (Desktop Application)
Suitable for users of the Claude Desktop application.
- Open the "Claude" desktop application
- Go to the "Settings" page and find the "Capabilities" section
- Find the "Skills" section and click "Upload skill"
- Select the downloaded
n8n-skills-{version}.zipfile to upload - After upload completes, you will see "n8n-skills" below. If not enabled, click to enable it.
- Return to the conversation window and ask questions about n8n. Successfully using n8n-skills indicates successful installation.
Basic Usage Examples
After installation, you can use it like this:
Query specific node information:
"What are the main features of the HTTP Request node?"
"How to send email using the Gmail node?"
"What programming languages does the Code node support?"
Explore workflow patterns:
"How to create a scheduled workflow?"
"What are the common node combinations for data transformation?"
"How to handle API errors and retries?"
Search for specific functionality:
"Which nodes can connect to Google Sheets?"
"What AI-related nodes are available?"
"What trigger nodes are available?"
Real-world Use Cases
Here are real-world scenarios using n8n Skills:
Case 1: AI-Assisted Workflow Design
Scenario: You want to create a workflow that "automatically sends daily operational reports."
Before n8n Skills: You need to manually search n8n documentation, research each node's functionality, and repeatedly trial-and-error to find the correct node combination.
After n8n Skills: Simply ask the AI assistant "How to create a scheduled email report workflow?", and the AI will recommend based on n8n Skills knowledge:
- Schedule Trigger (for scheduling)
- HTTP Request or Google Sheets (for data retrieval)
- Code or Item Lists (for data processing)
- Gmail or Send Email (for sending emails)
Time Saved: From hours of research reduced to minutes of design.
Case 2: Exploring the Best Node Combinations
Scenario: You want to integrate Slack notifications into your existing workflow but aren't sure which nodes are available.
With n8n Skills: Ask "Which nodes can send Slack messages?", and the AI will tell you:
- Slack node (official integration, most comprehensive features)
- HTTP Request (uses Slack API, highest flexibility)
- Webhook (receives Slack events)
Along with explaining the pros and cons of each node, suitable scenarios, and configuration methods.
Benefit: Quickly find the most suitable solution for your needs, avoiding detours.
Case 3: Learning Workflow Best Practices
Scenario: Beginners want to learn how to handle API errors and retry mechanisms.
With n8n Skills: Ask "How to handle HTTP Request errors and retries?", and the AI will reference 20 curated templates and 542 nodes of knowledge to share:
- How to use Error Trigger node
- Built-in retry settings for HTTP Request
- Error handling pattern using If node with Stop and Error
- Real-world workflow examples
Accelerated Learning: From zero foundation to mastering best practices, systematically improving skills.
FAQ
Skill won't load?
- Confirm the
SKILL.mdfile is in the correct location - Check if the file name is correct (case-sensitive)
- Verify the
resources/directory structure is complete - Restart the Claude application or refresh the webpage
File structure errors
- Ensure
SKILL.mdandresources/are in the same directory level - Do not modify the file structure inside the
resources/directory - If there are multiple directory levels after extraction, move the contents to the correct location
Version compatibility
- Newer versions of n8n may have new nodes, but most features will still work
- It's recommended to regularly update the Skill Pack to get the latest node information
How to update to a new version
- Download the latest version from GitHub Releases
- Delete the old Skill files
- Reinstall the new version following the installation steps
- Restart the Claude application
Reference Resources
Development Environment Setup
npm install
Development Commands
# Build project
npm run build
# Development mode
npm run dev
# Run tests
npm test
# Type checking
npm run typecheck
Technical Requirements
- Node.js >= 18.0.0
- TypeScript >= 5.3.0
Project Information
Maintainer
- Frank Chen (@haunchen)
- Project Homepage: https://github.com/haunchen/n8n-skills
Version Information
- Current Version: 1.2.0
- Supported n8n Version: v2.15.1
- Last Updated: April 2026
- Release Frequency: Follows n8n major version updates
Project Statistics
- Node Coverage: 542 built-in nodes
- Community Packages: 30+ packages
- Curated Templates: 20 templates
- Output Files: 126 files
- Total Documentation Size: 2.7 MB
- Supported Platforms: Claude Code, Claude.ai Web, Claude Desktop
Technical Features
- Multi-dimensional priority scoring system (usage frequency, documentation quality, community popularity)
- Node hierarchical merging strategy to optimize file structure
- Automated CI/CD pipeline ensuring quality
- Caching mechanism to accelerate build process
- Node compatibility matrix analysis
Acknowledgments
This project was built using the following resources:
- Based on the n8n-mcp project architecture created by Romuald Czlonkowski @ www.aiadvisors.pl/en
- Uses n8n node type definitions and metadata
- References n8n official documentation
Thanks to all contributors and the open source community for their support.
Related Links
- n8n Official Website: https://n8n.io
- n8n GitHub: https://github.com/n8n-io/n8n
- n8n Documentation: https://docs.n8n.io
- n8n-mcp Project: https://github.com/czlonkowski/n8n-mcp
Community and Support
We welcome your participation and feedback!
Issue Reporting and Feature Suggestions
- Encountered a problem? Please report it at GitHub Issues
- Have a feature suggestion? Feel free to open a new Issue for discussion
- Found a bug? Please provide detailed reproduction steps
Contributing
- Check CONTRIBUTING.md to learn how to contribute
- Fork the project and submit a Pull Request
- Improvements to documentation, bug fixes, and new features are all welcome
- Help translate into other languages
Stay Connected
- Follow GitHub Releases for notifications about new versions
- Star this project to support our development work
- Share with friends and colleagues who might need it
Technical Support
- Usage questions: Please first check the "FAQ" section
- Technical discussions: Feel free to initiate discussions in Issues
- n8n-related questions: Please refer to n8n Official Documentation or n8n Community Forum
License Information
This project is licensed under the MIT License, but includes the following third-party resources:
- n8n-mcp - n8n Model Context Protocol Integration
- License: MIT License
- Source: https://github.com/czlonkowski/n8n-mcp
- Author: Romuald Czlonkowski @ www.aiadvisors.pl/en
For detailed license information, please refer to ATTRIBUTIONS.md and LICENSE.
License Terms
MIT License - See LICENSE for details.
All trademarks and copyrights belong to their respective owners.
Get Started with n8n Skills
Ready to let AI assistants help you design n8n workflows?
Download Now
Visit GitHub Releases to download the latest version of n8n Skills and start experiencing AI-assisted workflow design!
Join the Community
- Star this project on GitHub to support our development
- Follow Releases for notifications about new versions
- Share your experience or suggestions at Issues
Contribute Your Ideas
n8n Skills is an open source project, and we welcome all forms of contribution:
- Report issues or suggest new features
- Improve documentation or translations
- Submit code improvements
- Share your use cases
Visit GitHub now and start your n8n automation skills journey!