Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

n8n Workflow Automation Skill · aiskillstore | DecimalAI

aiskillstore/n8n

SkillScore88Maintained

n8n workflow automation patterns and API integration. This skill should be used when creating n8n workflows, using webhooks, managing workflows via REST API, or integrating n8n with MCP servers. Covers workflow JSON structure, node patterns, and automation best practices.

DevOpsby aiskillstore

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

DevOps & SRE pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/aiskillstore-n8n/SKILL.md

decimalai skills pull aiskillstore-n8n --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/aiskillstore-n8n/SKILL.md https://app.decimal.ai/s/aiskillstore-n8n/SKILL.md

Raw file: /s/aiskillstore-n8n/SKILL.md ↗ · machine-readable summary at /s/aiskillstore-n8n.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

25 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 8d ago

Yes — +8% lift · at higher cost+139% tokens0% turns88% of whole cases pass with skill (22/25)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-16	✗→✓	▲ Improved	-13%	0%
case-23	✗→✓	▲ Improved	23%	0%
case-11	✓→✓	＝ Same ✓	408%	0%
case-01	✓→✓	＝ Same ✓	209%	0%
case-02	✓→✓	＝ Same ✓	92%	0%

Showing 5 of 25 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

aiskillstore / marketplace

skills/89jobrien/n8n/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1Bundled 2 files

Skill contents

references/

assets/

SKILL.md

n8n Workflow Automation Skill

This skill enables creating and managing n8n workflows for automation tasks.

Prerequisites

n8n instance running with API access:

bash
N8N_HOST=localhost
N8N_PORT=5678
N8N_API_KEY=your-api-key

Core Concepts

Workflow Structure

Every n8n workflow is JSON with this structure:

json
{
  "name": "Workflow Name",
  "nodes": [],
  "connections": {},
  "settings": {
    "executionOrder": "v1"
  }
}

Node Structure

Each node has:

json
{
  "id": "unique-id",
  "name": "Display Name",
  "type": "n8n-nodes-base.nodetype",
  "typeVersion": 1,
  "position": [x, y],
  "parameters": {},
  "credentials": {}
}

Connection Structure

Connections define data flow between nodes:

json
{
  "Source Node": {
    "main": [
      [{"node": "Target Node", "type": "main", "index": 0}]
    ]
  }
}

Common Workflow Patterns

1. Webhook-Triggered Workflow

Creates an HTTP endpoint that triggers workflow execution:

json
{
  "name": "Webhook Handler",
  "nodes": [
    {
      "id": "webhook",
      "name": "Webhook",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 2,
      "position": [250, 300],
      "webhookId": "my-webhook",
      "parameters": {
        "path": "my-endpoint",
        "httpMethod": "POST",
        "responseMode": "responseNode"
      }
    },
    {
      "id": "respond",
      "name": "Respond",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1.1,
      "position": [450, 300],
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ $json }}"
      }
    }
  ],
  "connections": {
    "Webhook": {
      "main": [[{"node": "Respond", "type": "main", "index": 0}]]
    }
  }
}

Access at: http://localhost:5678/webhook/my-endpoint

2. HTTP Request Pattern

Make external API calls:

json
{
  "id": "http",
  "name": "HTTP Request",
  "type": "n8n-nodes-base.httpRequest",
  "typeVersion": 4.2,
  "position": [450, 300],
  "parameters": {
    "method": "POST",
    "url": "https://api.example.com/endpoint",
    "authentication": "predefinedCredentialType",
    "nodeCredentialType": "myApiCredential",
    "sendBody": true,
    "specifyBody": "json",
    "jsonBody": "={{ JSON.stringify($json) }}"
  },
  "credentials": {
    "myApiCredential": {"id": "cred-id", "name": "My Credential"}
  }
}

3. Conditional Branching (IF Node)

Route data based on conditions:

json
{
  "id": "if",
  "name": "IF",
  "type": "n8n-nodes-base.if",
  "typeVersion": 2,
  "position": [450, 300],
  "parameters": {
    "conditions": {
      "options": {
        "caseSensitive": true,
        "leftValue": "",
        "typeValidation": "strict"
      },
      "conditions": [
        {
          "leftValue": "={{ $json.status }}",
          "rightValue": "success",
          "operator": {
            "type": "string",
            "operation": "equals"
          }
        }
      ],
      "combinator": "and"
    }
  }
}

4. Loop Over Items (SplitInBatches)

Process items in batches:

json
{
  "id": "batch",
  "name": "Loop Over Items",
  "type": "n8n-nodes-base.splitInBatches",
  "typeVersion": 3,
  "position": [450, 300],
  "parameters": {
    "batchSize": 10,
    "options": {}
  }
}

n8n REST API

List Workflows

bash
curl -s "http://localhost:5678/api/v1/workflows" \
  -H "X-N8N-API-KEY: $N8N_API_KEY"

Get Workflow

bash
curl -s "http://localhost:5678/api/v1/workflows/{id}" \
  -H "X-N8N-API-KEY: $N8N_API_KEY"

Create Workflow

bash
curl -s -X POST "http://localhost:5678/api/v1/workflows" \
  -H "X-N8N-API-KEY: $N8N_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "New Workflow", "nodes": [...], "connections": {...}}'

Update Workflow

bash
curl -s -X PUT "http://localhost:5678/api/v1/workflows/{id}" \
  -H "X-N8N-API-KEY: $N8N_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Updated", "nodes": [...], "connections": {...}}'

Activate/Deactivate

bash
curl -s -X POST "http://localhost:5678/api/v1/workflows/{id}/activate" \
  -H "X-N8N-API-KEY: $N8N_API_KEY"

curl -s -X POST "http://localhost:5678/api/v1/workflows/{id}/deactivate" \
  -H "X-N8N-API-KEY: $N8N_API_KEY"

Get Executions

bash
curl -s "http://localhost:5678/api/v1/executions?workflowId={id}&limit=10&includeData=true" \
  -H "X-N8N-API-KEY: $N8N_API_KEY"

Expression Syntax

n8n uses expressions for dynamic values:

| Syntax | Description | |--------|-------------| | ={{ $json.field }} | Access current item field | | ={{ $json.body.param }} | Access nested field | | ={{ $('Node Name').item.json.field }} | Access output from specific node | | ={{ $input.first().json }} | First input item | | ={{ $input.all() }} | All input items | | ={{ JSON.stringify($json) }} | Convert to JSON string |

Common Node Types

| Node | Type | Purpose | |------|------|---------| | Webhook | n8n-nodes-base.webhook | HTTP trigger | | HTTP Request | n8n-nodes-base.httpRequest | API calls | | Respond to Webhook | n8n-nodes-base.respondToWebhook | Return HTTP response | | IF | n8n-nodes-base.if | Conditional branching | | Switch | n8n-nodes-base.switch | Multi-way branching | | Set | n8n-nodes-base.set | Transform data | | Code | n8n-nodes-base.code | Custom JavaScript | | Split In Batches | n8n-nodes-base.splitInBatches | Loop processing | | Merge | n8n-nodes-base.merge | Combine branches |

MCP Integration

n8n can expose workflows as MCP tools via the built-in MCP server:

Enable MCP in workflow settings: "availableInMCP": true
Access MCP endpoint: http://localhost:5678/mcp-server/http
Use supergateway for Claude Code integration

Tips

Webhook Response: Use responseMode: "responseNode" with Respond node for control
Credentials: Store API keys in n8n credentials, reference by ID
Error Handling: Add Error Trigger node for failure notifications
Testing: Use /webhook-test/ path during development
ADF Format: Jira/Confluence require Atlassian Document Format for rich text

Want to see if aiskillstore/n8n would activate on your phrasing? Try it before you install — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/21/2026, 19:50 UTC · 25 cases

Lift

+8%

88% vs 80%

Agent turns

1.0 vs 1.0

Tokens

+139%

2,955.92 vs 1,235.48

fail → pass · 2pass → pass · 20fail → fail · 3

Per-case results (25)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-11 Provide the exact n8n node type string used for looping over an incoming item collection in batches. Avoid general names like 'n8n-nodes-base.loop' or 'n8n-nodes-base.batchProcessor'.	pass→pass	2,824	1,920	-32%	1	1	0%	478	2,428	+408%	0	0	—
case-01 Construct a curl command to query the list of active workflows from an n8n REST API instance at http://localhost:5678. Pass the API key string stored in the environment variable MY_KEY. Standard REST APIs usually accept 'Authorization: Bearer <token>', but supply the exact HTTP header required by n8n REST API endpoints.	pass→pass	4,406	3,946	-10%	1	1	0%	917	2,832	+209%	0	0	—
case-02 Write an HTTP API call using curl to activate an existing n8n workflow with ID 'wf_999' on a local instance at http://localhost:5678. A common mistake is sending a PATCH request to /api/v1/workflows/wf_999 with a JSON body, but specify the exact dedicated endpoint route and HTTP method used by n8n.	pass→pass	6,983	2,993	-57%	1	1	0%	1,410	2,711	+92%	0	0	—
case-03 Write a curl command to fetch past execution history logs for a specific n8n workflow with ID 'wf_123' from the n8n REST API, limiting output to 10 entries and including full execution data. Avoid targeting /api/v1/workflows/wf_123/executions.	pass→pass	4,791	3,401	-29%	1	1	0%	949	2,763	+191%	0	0	—
case-04 In an n8n Set node expression field, write an expression to extract the 'user_email' property from the currently processed incoming JSON item. Resist using handlebars like '{{ $json.user_email }}' or template literals like '${json.user_email}'.	fail→fail	3,619	7,373	+104%	1	1	0%	570	3,372	+492%	0	0	—
case-05 Construct an n8n expression to extract the 'order_id' field from the item output JSON of an earlier node explicitly named 'Customer Lookup'. Avoid standard array index references like '$nodes["Customer Lookup"].data.order_id'.	pass→pass	4,871	3,020	-38%	1	1	0%	804	2,664	+231%	0	0	—
case-06 Define a JSON node object for an n8n trigger node that handles incoming HTTP POST webhooks. Specify the exact internal node type string, avoiding generic names like 'n8n-nodes-base.webhookTrigger' or 'n8n-nodes-base.httpWebhook'.	pass→pass	6,265	4,858	-22%	1	1	0%	1,181	3,020	+156%	0	0	—
case-12 Write an n8n expression to access the JSON content of the first item from the incoming input array entering a node. Avoid using zero-indexed array syntax like '$input[0]'.	pass→pass	5,232	8,516	+63%	1	1	0%	874	2,685	+207%	0	0	—
case-07 Construct the JSON definition for an n8n node that makes outbound REST API calls. Provide the official n8n node type string, avoiding alternatives like 'n8n-nodes-base.http' or 'n8n-nodes-base.rest'.	pass→pass	5,640	4,336	-23%	1	1	0%	1,106	2,983	+170%	0	0	—
case-08 Specify the exact node type identifier for an n8n node that returns a custom JSON HTTP response back to an incoming Webhook caller. Avoid standard guesses like 'n8n-nodes-base.webhookResponse' or 'n8n-nodes-base.httpResponse'.	pass→pass	4,330	3,169	-27%	1	1	0%	762	2,692	+253%	0	0	—
case-09 When configuring an n8n Webhook node to delegate its HTTP response body and status code to a downstream Respond node, what value must be assigned to the Webhook node's 'responseMode' parameter? Do not use 'onReceived' or 'lastNode'.	pass→pass	4,355	2,528	-42%	1	1	0%	618	2,596	+320%	0	0	—
case-10 During workflow building and manual testing in n8n, test webhook calls use a specific URL path segment rather than production webhook paths. What is the URL path prefix used specifically for test webhook executions?	pass→pass	3,365	2,384	-29%	1	1	0%	524	2,498	+377%	0	0	—
case-13 Write an n8n expression to retrieve all incoming input items entering the current node as a complete array. Avoid property syntax like '$input.items' or '$input.data'.	pass→pass	4,465	3,972	-11%	1	1	0%	747	2,774	+271%	0	0	—
case-14 Inside an n8n workflow JSON object under the top-level 'settings' key, which execution ordering version string should be specified for modern workflow execution? Avoid setting executionOrder to 'v2' or 'legacy'.	pass→pass	3,842	2,216	-42%	1	1	0%	662	2,521	+281%	0	0	—
case-15 To expose an n8n workflow as a callable tool for Model Context Protocol (MCP) clients, what boolean setting key must be included in the workflow settings JSON object?	pass→pass	6,418	1,623	-75%	1	1	0%	1,120	2,363	+111%	0	0	—
case-16 What default HTTP URL path endpoint on an n8n server instance serves the built-in Model Context Protocol (MCP) server interface?	fail→pass	14,913	1,948	-87%	1	1	0%	2,737	2,389	-13%	0	0	—
case-17 In an n8n workflow JSON definition, show how a connection linking a source node named 'Webhook' to a target node named 'Respond' is structured within the connections object.	pass→pass	4,865	4,320	-11%	1	1	0%	891	2,864	+221%	0	0	—
case-18 Provide the exact node type string used for conditional binary branching in n8n workflows. Avoid common flow control names like 'n8n-nodes-base.condition' or 'n8n-nodes-base.filter'.	pass→pass	3,016	2,166	-28%	1	1	0%	462	2,492	+439%	0	0	—
case-19 Write a curl command targeting an n8n REST API instance at http://localhost:5678 to deactivate an active workflow with ID 'wf_555'. Avoid using DELETE or PUT HTTP methods.	pass→pass	6,312	2,768	-56%	1	1	0%	1,248	2,641	+112%	0	0	—
case-20 When sending formatted rich text to Jira or Confluence nodes in automation workflows, plain markdown strings often break formatting. What specific document formatting standard is required by Atlassian node parameters?	pass→pass	8,710	3,851	-56%	1	1	0%	1,394	2,790	+100%	0	0	—
case-21 Provide the official n8n node type identifier used for running custom JavaScript or Python scripts inside a workflow node. Do not use 'n8n-nodes-base.function' or 'n8n-nodes-base.script'.	pass→pass	2,494	2,510	+1%	1	1	0%	396	2,466	+523%	0	0	—
case-22 Provide the exact node type identifier for multi-path conditional branching in n8n. Avoid generic labels like 'n8n-nodes-base.router' or 'n8n-nodes-base.case'.	pass→pass	4,262	6,606	+55%	1	1	0%	682	2,455	+260%	0	0	—
case-23 How do I configure PostgreSQL database connection pool sizes and read-replica connection strings directly inside an n8n workflow JSON settings block?	fail→pass	17,575	9,775	-44%	1	1	0%	3,103	3,807	+23%	0	0	—
case-24 Write the full TypeScript source code for a custom n8n community node npm package implementing the INodeType class interface and build configuration.	fail→fail	19,424	20,311	+5%	1	1	0%	3,888	6,423	+65%	0	0	—
case-25 Write a Zapier CLI creates trigger schema module using zapier-platform-cli for posting JSON records to an external service.	fail→fail	17,117	13,014	-24%	1	1	0%	3,364	4,669	+39%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted. The headline lift of +8 percentage points is the difference between those two pass rates over the 25 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 80%
Pass rate with the skill: 88%
Cases attempted: 25
Cases counted toward lift: 25
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/21/2026, 19:50 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-16

Prompt

What default HTTP URL path endpoint on an n8n server instance serves the built-in Model Context Protocol (MCP) server interface?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run88

Live evalNeeds ≥20 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+209% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 7 days ago

Other measured skills in the registry, with their headline benchmark lift.