using-n8n-skills
Skill1 starsChanged 3 months agoMITView on GitHub
Use when working with n8n workflows in any capacity. Always-on meta-skill and router for the Molavi n8n-agent-skills pack. Routes to capability skills, summarizes MCP tool families across official and community servers, and lists non-negotiable cross-cutting rules.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- LucasGS520/Claude-Skillstheir own version, first seen Sep 15, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- LucasGS520/Claude-Skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- llms.txtllms.txt
- n8n-agentsSkill
- n8n-ai-seo-geo-automationSkill
- n8n-binary-dataSkill
- n8n-code-nodesSkill
- n8n-credentials-securitySkill
- n8n-data-tablesSkill
- n8n-debuggingSkill
- n8n-error-handlingSkill
- n8n-expressionsSkill
- n8n-loopsSkill
- n8n-mcp-toolingSkill
- n8n-node-configurationSkill
- n8n-production-opsSkill
- n8n-subworkflowsSkill
- n8n-workflow-lifecycleSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 92
Never claim production-ready without stages 4–5.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Using n8n Skills (Molavi pack)
Maintained by Taghi Molavi · GitHub
This meta-skill is the router. Capability skills hold the real rules. Load this first on every n8n task, then invoke the matching skill via the Skill tool before writing SDK/JSON, configuring nodes, or calling MCP tools.
Non-negotiables
- Invoke the relevant skill before any n8n action — not only MCP calls. Designing, wiring, expressions, agents, errors, credentials: load the skill first.
- Validate AND verify before publish/activate — lint/validate is necessary but not sufficient. Re-fetch the workflow and inspect
connections. Silent dropped wires are common. - Secrets never in text fields — use n8n credentials. For HTTP Request, bind an official credential type. Never paste API keys into parameters, sticky notes, or Code nodes.
- Prefer live schema over training memory — n8n parameters, typeVersions, and MCP tool names drift. Trust
get_node_types/ node docs / lint output over recollection. - Ask before side effects —
test_workflow,execute_workflow, and activate can send Slack messages, write DBs, charge cards. Confirm with the user when unpinned nodes have real side effects.
Strong defaults
| Default | Detail | Skill |
| Code node is last resort | Expression → Edit Fields arrow fn → Code | n8n-code-nodes |
| Reuse = sub-workflow | Search tagged subworkflow before building | n8n-subworkflows |
$('Node').item.json.x over bare $json | Stable across refactors and merges | n8n-expressions |
| Error branch on fallible nodes | Webhooks need correct 4xx/5xx shapes | n8n-error-handling |
| Lint before create/activate | Catch typeVersion, AI topology, IF v1, webhooks | n8n-mcp-tooling |
| Diagnose with execution tools | Silent 0-item drops are #1 pain | n8n-debugging |
MCP landscape (pick what the user has)
This pack is MCP-server agnostic. Detect which tools exist in the session and adapt names.
| Family | Typical install | Strength |
| Official n8n instance MCP | n8n 2.2.0+ · Settings → Instance-level MCP · OAuth HTTP | Native create/update/validate/publish, folders, SDK |
AutomateLab @automatelab/n8n-mcp | npx -y @automatelab/n8n-mcp + API key | workflow_generate, workflow_lint, execution_explain, AI agent topology |
| czlonkowski/n8n-mcp | npx / Docker + API key | Huge node corpus, templates, docs search |
| salacoste workflow-builder | npm package + multi-instance | Conversational CRUD, multi-env |
| leonardsellem/n8n-mcp-server | npm/Docker + API key | Classic workflow/execution bridge |
See n8n-mcp-tooling for routing tables and config snippets. See configs/ for ready-to-paste MCP JSON.
Skill index
| Skill | When to invoke |
n8n-workflow-lifecycle | Start/design/finish a workflow; naming; folders; validate→test→publish |
n8n-mcp-tooling | Choosing/configuring MCP servers; tool routing; multi-server setups |
n8n-expressions | {{}}, $json, $node, Luxon, expression errors |
n8n-node-configuration | Any node parameters, triggers, webhooks, Merge, Switch, HTTP, DB |
n8n-code-nodes | Code node decision tree + JS patterns |
n8n-loops | Batching, pagination, Loop Over Items vs default per-item |
n8n-agents | LangChain Agent, tools, memory, RAG, structured output |
n8n-error-handling | Production, webhooks, error workflows, response shapes |
n8n-credentials-security | Auth, API keys, credential hygiene |
n8n-subworkflows | Reuse, tags, agent-as-tool, passthrough vs Define Below |
n8n-binary-data | Files, images, attachments, vision, binary merge |
n8n-data-tables | Data Tables schema, dedup, persistent state |
n8n-debugging | Failures, 0 items, unresolved expressions, execution explain |
n8n-ai-seo-geo-automation | SEO/AEO/GEO pipelines, content ops, AI visibility workflows |
n8n-production-ops | Multi-env, backup, CI lint, activation policy, observability |
Red flags → STOP and invoke
| Thought | Action |
| "Simple workflow, just build it" | n8n-workflow-lifecycle |
| "Add a Set node to map one field" | n8n-expressions (usually inline) |
| "Code node is easier" | n8n-code-nodes |
| "Validation passed → publish" | Lifecycle checklist + re-fetch connections |
| "Agent looks wired" | n8n-agents tool name/description quality |
| "Loop Over Items for each row" | n8n-loops (default iteration often enough) |
| "Paste the API key in the header" | n8n-credentials-security |
| "Why is this empty?" | n8n-debugging + execution explain |
Session protocol
- Confirm n8n goal (build / fix / audit / automate SEO-GEO / ops).
- Detect available MCP tools; if none, offer config from
configs/or guide JSON-only + lint path. - Load this router + the 1–3 matching capability skills.
- Plan → build → lint/validate → verify connections → test (with consent) → activate/publish → handoff notes.
- Never claim production-ready without stages 4–5.
Related Molavi projects
- molavi.pro — Taghi Molavi
- geo-scope — GEO / AI visibility benchmark
- mcp-geo-server — GEO audit MCP
- mcp-agent-skills-hub — curated agent skills
- lean-agent-skills — token-efficient skills installer
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.