How to Use AI to Draft Customer Response Emails

Have AI prepare response drafts for your support team to review and send.

What This Integration Does

This is the human-in-the-loop cousin of full auto-reply. Instead of letting Spojit send messages directly, this workflow watches your support inbox, drafts a reply grounded in a Knowledge collection, and posts the draft somewhere a human can review, edit, and send. Agents go from cold-typing every reply to polishing a near-final draft, which typically halves handle time.

The workflow runs per inbound conversation. For each new message it retrieves Knowledge snippets, drafts a reply with a Connector node in Agent mode, and posts a preview into Slack and a tag onto the Front conversation so a human can review and send. Nothing goes out automatically: the reply is only sent after a person acts on it.

Prerequisites

A Front connection (or equivalent inbox connector) with permission to read conversations and post messages.
A Slack connection if you want drafts surfaced in a channel for review.
A Knowledge collection containing your support docs, FAQs, and policy pages.

Step 1: Trigger on New Inbound Message

Drop a Trigger node onto the canvas and set its type to Webhook. Wire Front Rules to POST inbound messages from your support inbox to the webhook. The payload carries conversationId, messageId, sender details, subject, and body.

Step 2: Knowledge - Pull Supporting Docs

Add a Knowledge node in Query mode. Set Collection to your persistent support-docs collection and pass {{ trigger.body }} as the Prompt. Leave Result Count at 5 so the answer stays focused on the most relevant material, and set an Output Variable of knowledge to reference downstream.

Step 3: Draft the Response in Agent Mode

Add a Connector node and switch it to Agent mode so the AI can reason over the inbound message and the retrieved knowledge. Use a prompt like:

You are a Spojit support agent. Draft a reply to the customer
below using ONLY the knowledge provided. Be friendly, concise, and
accurate. If the knowledge does not answer the question, draft a short
holding reply that says we'll get back to them and flag the gap.

Customer email:
{{ trigger.body }}

Knowledge:
{{ knowledge }}

Define a Response Schema with draft (the reply body) and gap (true if it could not answer from the knowledge) so reviewers can prioritize the tough ones, and set an Output Variable of agent. Agent mode uses AI credits per run.

Step 4: Tag the Conversation for Review

Add a Connector node pointing at the Front connector in Direct mode and pick the add-tag tool. Tag the inbound conversation so reviewers can find drafted items quickly:

conversationId: {{ trigger.conversationId }}
tag: ai-draft-ready

The conversation stays in your support inbox unread. Because Front does not expose a draft-creation tool, the proposed reply itself is surfaced to a human in Slack in the next step, where they can copy, edit, and send it from Front.

Step 5: Deliver the Draft to Reviewers in Slack

Add a Connector node pointing at the Slack connector in Direct mode and call the send-message tool into your support channel. Include the customer's subject, the full draft {{ agent.draft }}, and a link back to the Front conversation. When {{ agent.gap }} is true, prepend a "needs human research" note so it jumps the queue.

Step 6: Require Sign-Off Before High-Stakes Replies (Optional)

For sensitive queues you can gate the draft behind a real approval. Add a Human node before the Slack step, set its Message to a summary of the draft, and add an Approval slot for your support lead. The workflow pauses until the slot is satisfied in the Approvals inbox; an APPROVED outcome continues to Slack, while a rejection halts the run so nothing reaches the customer.

Tips

Always have the AI return short drafts. Reviewers edit ten-line drafts faster than they edit thirty-line ones.
Include the Knowledge answer the agent drafted from in the Slack notification. Reviewers can spot-check accuracy at a glance.
Track edit distance between the AI draft and what actually got sent. If it's huge, your Knowledge collection is underpowered.

Common Pitfalls

Drafts that contradict policy because the Knowledge collection has duplicate or outdated docs. Curate it aggressively.
Letting drafts sit. Add an SLA escalation: if a draft is older than 30 minutes, ping a backup reviewer.
Drafting on noise. Filter out auto-responders and notifications at the trigger so reviewers don't waste cycles on junk drafts.

Testing

Pick three real recent tickets your team already answered. Run them through the workflow and compare the AI draft to what your team actually sent. Adjust the prompt and Knowledge collection until the drafts feel like a 70% starting point. Then enable it on one inbox.