AI Readiness Auditor Agent for n8n
Walk through how your business actually works and get an honest list of what AI would help with, what it would not, and what to do first.
Agent contents
- 01ai-readiness-auditor-agent.pdf
One indexed handbook: what it does, how to set it up, the workflow node by node, the rules it follows, the prompt, its limits, troubleshooting, and a real recorded run.
- 02ai-readiness-auditor-agent.n8n.json
The live workflow. Import this into n8n.
- 03ai-readiness-auditor-agent-smoke-test.n8n.json
A credential-free run that proves the gates work before you spend anything.
- 04INPUT_SCHEMA.json
What the workflow accepts.
Inside
What’s inside
Everything included in the downloadable agent workflow.
One indexed handbook: what it does, how to set it up, the workflow node by node, the rules it follows, the prompt, its limits, troubleshooting, and a real recorded run.
The live workflow. Import this into n8n.
A credential-free run that proves the gates work before you spend anything.
What the workflow accepts.
What it returns, field by field.
Acceptance cases, including the ones designed to fail.
Audience
Who it is for
Use it when these situations match your work.
- Consultants running discovery and writing proposals
- Agencies qualifying which clients are ready to automate
- Builders validating an automation opportunity before quoting
- n8n users who want a tested starting workflow
Usage
How to use it
A practical sequence from download to finished output.
- Import and run the credential-free smoke test.
- Import the live workflow and connect one OpenAI credential.
- Fill in the Set Run Context node with your own details.
- Run the built-in sample and check every fact against the input.
- Paste your own input and review the result before acting on it.
AI Readiness Auditor Agent for n8n FAQ
Will it give me an ROI figure I can quote?
No, and deliberately. Hour estimates come only from what the client actually said. If no figure was given it returns 'not stated' rather than modelling a saving from an assumed baseline. An invented ROI is the fastest way to lose a client's trust.
Can I put the output straight in front of a client?
Review it first. requires_human_review defaults to true because the scoring rests on interview claims, not measurement. The structure is client-ready; the numbers need your judgement.
Why was an enthusiastic client scored NOT_READY?
Enthusiasm is not readiness. NOT_READY means no documented process and no named owner. Automating an undocumented workflow just encodes the chaos, which is a genuinely useful thing to tell a client early.
Does it suggest work I cannot deliver?
It should not. Opportunities are constrained to the services you list in the run context, so it will not propose an integration you do not build.
Is this an actual n8n workflow?
Yes. The download contains an importable live workflow plus a separate credential-free smoke test. It is not a prompt to paste into a chat window.
What do I need to set up?
One OpenAI credential in n8n. There is no OAuth and no external service to connect, because you paste the input directly into the workflow.
Can I use this for client work?
Yes. Review outputs before acting on them and do not resell the kit as-is.
Which AI model should I use?
The workflow ships with an OpenAI chat model selected in n8n. You can change it, but rerun the acceptance cases in TEST_CASES.json before relying on another model.
Does it act on the output by itself?
No. The workflow ends at a reviewed result. It does not send, post, file, or write to any external system.
Custom automation
Want this agent connected to your workflow?
I can connect this agent workflow to your inbox, documents, sheets, CRM, or internal tools, with review steps and safeguards built in.
Build my workflow →