I built an n8n workflow that builds, tests, and deploys other n8n workflows from chat

Description and use

An automated n8n pipeline that generates, validates, tests, and deploys workflows from chat descriptions, ensuring correctness and safety.

Problem

LLMs often generate invalid n8n workflow JSON, mixing node versions and wiring errors, making deployment unreliable.

Outcome

The workflow reliably generates valid n8n workflows from chat, validates them, runs tests, and deploys them while preventing credential leakage and unsafe nodes.

Workflow

• 1. User sends a one‑sentence automation request to the Chat Trigger.
• 2. Claude (Spec model) converts the sentence into a spec (monitor or blueprint).
• 3. Claude (Build model) generates workflow JSON using pinned node types, versions, and fields.
• 4. Validation node checks node types, fields, wiring, expressions, and references; auto‑repairs safe mistakes and loops up to 3 attempts.
• 5. Testing/Review node runs hidden test cases for monitors or a second Claude review for blueprints.
• 6. Deployment node creates the workflow via n8n API, activates monitors, and saves blueprints inactive with a Setup note.
• 7. User receives chat reply with test results, workflow link, and next steps.

Evidence

Getting an LLM to write n8n workflow JSON is easy. Getting JSON that imports cleanly, uses node versions that exist, and does what you asked is harder.

Source

You describe an automation in one sentence in chat, and the workflow builds it, checks it, tests or reviews it, and saves it to your n8n instance.

Source

Built workflows never contain credentials.

Source