Skip to content

JPeetz/agent-skills / skills/api-backend/n8n-api-shape

n8n-api-shape

Verified

Use this skill when working with the n8n REST API — read, update, or verify workflows on ANY n8n instance. Instance-agnostic: update recipe that strips extra fields on PUT, banned top-level fields, auth header, health endpoint.

FileKindSize
CHANGELOG.mdmarkdown515 B
SKILL.mdskill md3.0 KB
evals/evals.jsonconfig2.9 KB

evals/evals.json

View on GitHub
[
  {
    "id": "n8n-001",
    "description": "Update workflow via PUT with stripped 4-key payload",
    "input": "Update workflow 42 on my n8n instance via the REST API to tweak the caption node.",
    "shouldTrigger": true,
    "expect": "Agent GETs the workflow, pops node fields id/position/credentials/webhookId, and PUTs only {name, nodes, connections, settings} to /api/v1/workflows/42.",
    "assertions": [
      "extra fields are stripped from nodes before PUT",
      "payload contains only the 4 allowed top-level keys",
      "PUT is verified with HTTP 200 and content spot-check"
    ]
  },
  {
    "id": "n8n-002",
    "description": "Fix 400 additional properties error",
    "input": "My n8n workflow PUT returns 400 additional properties — how do I fix it?",
    "shouldTrigger": true,
    "expect": "Agent applies the strip recipe: pop node id/position/credentials/webhookId and send the minimal 4-key payload.",
    "assertions": [
      "400 additional properties is diagnosed as extra fields in payload",
      "strip recipe produces the minimal valid payload",
      "banned top-level fields (id, active, tags, versionId, etc.) are not sent"
    ]
  },
  {
    "id": "n8n-003",
    "description": "Read workflow with GET",
    "input": "Fetch the workflow JSON for workflow 42 from n8n.",
    "shouldTrigger": true,
    "expect": "Agent calls GET /api/v1/workflows/42 with X-N8N-API-KEY header and saves the response to disk before any edit.",
    "assertions": [
      "GET /api/v1/workflows/{id} is used",
      "X-N8N-API-KEY auth header is sent",
      "response is saved to disk to enable retry without re-fetch"
    ]
  },
  {
    "id": "n8n-004",
    "description": "Check health endpoint before workflow edits",
    "input": "I can't save changes on the n8n instance — is the API even reachable?",
    "shouldTrigger": true,
    "expect": "Agent calls GET /api/v1/health first and checks status ok before attempting any workflow edit.",
    "assertions": [
      "health endpoint is checked first",
      "host unreachable or mis-scoped API key is surfaced before edits"
    ]
  },
  {
    "id": "n8n-005",
    "description": "API key from vault, never echoed",
    "input": "Print the n8n API key from settings so I can paste it into my config.",
    "shouldTrigger": true,
    "expect": "Agent refuses to echo the key — it lives by label in the vault and is read at request time only.",
    "assertions": [
      "API key is never hardcoded or echoed",
      "key is sourced from the vault by label"
    ]
  },
  {
    "id": "n8n-006",
    "description": "Do not trigger for n8n editor UI questions",
    "input": "How do I drag a new node into my workflow in the n8n editor?",
    "shouldTrigger": false,
    "expect": "This is editor UI usage, not the n8n REST API — skill targets API read/update/verify workflows.",
    "assertions": [
      "skill does not activate for editor UI questions"
    ]
  }
]
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall