Skip to content
NEWMarketplace and your team: sell what you build, and get plans and changes that are tested before you see them. See what is new

Library / Customer Support & Helpdesk

Customer Support & Helpdesk

Draft guarded review replies with OpenAI and escalate risky ones to Slack

Triggered by a webhook carrying a review, it classifies the review and drafts a reply. Guardrails log safe drafts as ready to post and send risky cases to the owner on Slack.

What it does

## AI Review Reply Agent with Guardrails POST a review → get a brand-safe reply draft. Flow: validate input → LLM classifies + drafts → code guardrails (length, blocked phrases, low rating, model flag) → safe replies are logged as ready_to_post, risky ones go to the owner on Slack. Nothing is posted publicly by this workflow. Body: {business, platform, author, rating, review, policy, id}

The steps

  1. Webhook
  2. Config
  3. Validate Input
  4. Valid?
  5. Respond 400
  6. Build Prompt
  7. LLM
  8. Parse + Guardrails
  9. Respond 200
  10. Escalate?
  11. Alert Owner (Slack)
  12. Log Approved Reply

Similar automations