# AUTOMATION: AI Fit Triage

**Automation stage:** Automation-Ready — this workflow is designed for use with Make, n8n, Zapier, or MCP.
**Manual version:** Run each step manually first to confirm outputs before connecting automation.

---

## Trigger
**Phrase:** `Run automation: AI Fit Triage`
**Trigger type:** Manual / scheduled / webhook (configure in your automation tool)

## Input Schema
```
{
"context": "[describe what this automation receives as input]",
"goal": "[what a successful output looks like]",
"format": "[output format required by the next step or tool]"
}
```

## Core Prompt
ROLE:
You are an AI triage advisor.

GOAL:
Evaluate a workflow and decide whether the right answer is AI automation, templates, delegation, process change, or no intervention at all.

INPUT:
Workflow and problem: [WHAT IS HAPPENING AND WHAT FEELS BROKEN]
Current method: [HOW IT IS DONE TODAY]
Constraints and risks: [BUDGET, ACCURACY NEEDS, COMPLIANCE, HUMAN JUDGEMENT]

TASKS:
1. Diagnose the core problem in the workflow.
2. Evaluate whether AI is actually a good fit.
3. Compare AI against non-AI alternatives such as templates, better process, delegation, or tooling cleanup.
4. Recommend the best path and why.
5. If AI is appropriate, define the smallest useful pilot.

CONSTRAINTS:
- Wait for user data before starting.
- Do not assume AI is the answer.
- Do not invent missing constraints.
- Prioritise practical fit, accuracy, and risk.

OUTPUT FORMAT:
- Core problem diagnosis
- Best intervention type
- Why this is the right choice
- Smallest useful next step

## Output Schema
```
{
"output": "[structured result]",
"status": "complete | needs_review | failed",
"issues": "[list any problems found — empty if none]",
"handoff": "[summary for the next step or tool]"
}
```

## Human Approval Checkpoint
⚠️ **Require human review before:**
- Sending to external recipients
- Publishing or posting publicly
- Making financial or legal decisions
- Updating production systems

## Error Handling
- If input is missing required fields: return `status: failed` with list of missing fields
- If output quality is uncertain: return `status: needs_review` with specific reason
- Do not guess missing inputs — fail cleanly with a clear message

## Platform Notes
- **Make / n8n:** Map input fields to the schema above. Add a human approval module before output delivery.
- **Zapier:** Use a Filter step to check `status === complete` before continuing the Zap.
- **MCP:** Register trigger phrase as a tool. Validate input schema before invoking.