← 返回 Skills 市场
billzhuang6569

n8n Master

作者 Bill Zhuang · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ⚠ suspicious
183
总下载
0
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install n8n-master
功能描述
Answers n8n questions and builds n8n workflows, node configurations, Code node JavaScript, HTTP Request setups, API integrations, toolbox-assisted API tests,...
使用说明 (SKILL.md)

n8n Master

Purpose

Use this skill to answer n8n questions and produce practical n8n artifacts: workflow designs, node configurations, Code node JavaScript, HTTP Request setups, external API integrations, and importable workflow JSON.

Do not invent node parameters, API fields, permissions, or response paths. Prefer local references first, then official source files. If the references do not support an answer, say what is missing.

Routing

Start with the narrowest relevant index:

  • n8n concepts: references/wiki/index/ALL-Concepts.md, then references/wiki/concepts/.
  • n8n nodes: references/wiki/index/ALL-Nodes.md, then references/wiki/nodes/.
  • External APIs: references/wiki/index/ALL-APIs.md, then references/wiki/api-cards/.
  • Workflow patterns: references/wiki/index/ALL-Recipes.md, then references/wiki/recipes/.
  • Useful Q&A: references/wiki/index/ALL-QAs.md, then references/wiki/qa/.
  • Source index: references/wiki/index/ALL-Sources.md. This release is wiki-only and does not bundle raw source corpora.
  • Document compilation: references/compiler/ai-document-compiler.md when the user asks to absorb, compile, update, or learn from docs.
  • Toolbox operations: references/toolbox/ only when the user explicitly asks to test an API, inspect/create Feishu Base fields, or ingest online docs.

If an index is missing or incomplete, use rg over references/wiki/. If the user adds new raw docs later, place them under references/source/ and run the AI compiler workflow.

Answer Modes

Concept or Debugging Question

  1. Search the relevant index and card.
  2. Check source files for fragile details.
  3. Answer with source file paths.
  4. Separate confirmed facts from inference.

HTTP Request Node

Use references/templates/http-node-output.md.

Include method, URL, auth, headers, query params, body, response paths, and common failure checks. Do not include real secrets. Use credential names or environment variable placeholders.

Code Node

Use references/templates/code-node-output.md.

State the assumed input item structure and execution mode. Return valid n8n items, usually an array of { json: ... } objects for "Run Once for All Items". Avoid unsupported globals unless a local source confirms them.

Workflow JSON

Use references/templates/workflow-json-output.md.

Only output full workflow JSON when the user asks for importable/copy-paste workflow JSON. Keep active: false. Use placeholder credential names. Mention that credentials must be bound in n8n after import.

Production Hardening

Use hardening recipes when the user asks for reliability, auditability, retries, idempotency, human review, safe reruns, or "no silent failure".

Add only the hardening needed for the workflow:

  • dedup key and idempotent writes
  • run_id and status logging
  • retries/backoff for safe operations
  • final failure notification
  • human review queue for risky failures
  • runbook when the workflow is meant to operate long-term

Toolbox

Toolbox scripts are not part of normal answering. Use them only when explicitly helpful:

  • scripts/toolbox/api_tester.py: lightweight Postman-like API test and n8n HTTP Request draft.
  • scripts/toolbox/feishu_get_bitable_schema.py: read Feishu Base field schema.
  • scripts/toolbox/feishu_create_bitable_fields.py: create Feishu Base fields; default dry-run.
  • scripts/toolbox/firecrawl_ingest_docs.py: ingest online docs into references/source/.

Before running a toolbox script that may touch the network, preserve the local proxy environment. Before write operations, dry-run first unless the user clearly requests execution.

Document Compilation

When the user adds new docs or asks this skill to learn from source files, do not rely on a script to generate final knowledge. Use references/compiler/ai-document-compiler.md.

The AI compiler must read source files, classify them, write compact wiki cards, update references/wiki/index/ALL-*.md, and append to references/wiki/index/Compilation-Log.md. Scripts may only help with mechanical inventory and hashes.

Source And Safety Rules

  • Never write real tokens, cookies, app secrets, or API keys into outputs, workflow JSON, references, source files, logs, or examples.
  • For Feishu tools, use FEISHU_APP_ID and FEISHU_APP_SECRET from the environment; never print them.
  • For Firecrawl, use FIRECRAWL_API_KEY from the environment; never print it.
  • For writes, deletes, batch updates, external messages, or customer/financial data, prefer dry-run and explicit confirmation.
  • If source documents disagree, cite the conflict and avoid pretending it is resolved.
  • Keep answers copy-paste practical; avoid long tutorials unless the user asks.
安全使用建议
Before installing, confirm you want a broad n8n workflow/API helper. When using toolbox features, run dry-runs first, inspect generated requests, use least-privilege credentials, keep real secrets out of prompts and workflow JSON, and review any imported workflow before activating it. ClawScan detected prompt-injection indicators (unicode-control-chars), so this skill requires review even though the model response was benign.
能力标签
cryptorequires-walletcan-make-purchasesrequires-oauth-tokenrequires-sensitive-credentials
能力评估
Purpose & Capability
The stated purpose, bundled n8n reference corpus, workflow templates, and toolbox documentation are coherent for answering n8n questions and building workflows. The skill covers broad API integrations, including mutating API patterns, so users should review generated actions before using them.
Instruction Scope
Instructions emphasize local references, source-grounded answers, inactive workflow JSON, placeholder credentials, dry-runs, and explicit confirmation for writes/deletes. The document compiler can update persistent reference files when the user asks the skill to learn from new docs.
Install Mechanism
There is no install spec or automatic setup, but helper scripts are included and the registry lists no upstream source or homepage. Static scan was clean, but provenance is limited.
Credentials
No required environment variables are declared, but optional Feishu, Firecrawl, and n8n/API credentials are described for toolbox/API workflows. This is purpose-aligned, with guidance not to print or embed secrets.
Persistence & Privilege
No background agent or auto-start persistence is shown. Persistent effects are limited to user-directed reference/source/wiki file updates during document ingestion/compilation and external API writes when explicitly executed.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install n8n-master
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /n8n-master 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.2
Wiki-only release: remove raw source corpora while keeping compiled nodes, API cards, concepts, recipes, QA cards, indexes, compiler prompt, templates, and toolbox helpers.
v1.0.1
Sanitize compact API index wording to avoid false-positive secret detection; keep compact routing, indexes, API cards, compiler prompt, templates, and toolbox helpers.
v1.0.0
Initial compact registry release: n8n-master routing, indexes, API cards, short node/concept/recipe cards, Feishu API pack metadata, compiler prompt, templates, and toolbox helpers.
元数据
Slug n8n-master
版本 1.0.2
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 3
常见问题

n8n Master 是什么?

Answers n8n questions and builds n8n workflows, node configurations, Code node JavaScript, HTTP Request setups, API integrations, toolbox-assisted API tests,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 183 次。

如何安装 n8n Master?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install n8n-master」即可一键安装,无需额外配置。

n8n Master 是免费的吗?

是的,n8n Master 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

n8n Master 支持哪些平台?

n8n Master 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 n8n Master?

由 Bill Zhuang(@billzhuang6569)开发并维护,当前版本 v1.0.2。

💬 留言讨论