Classify Intercom Messages & Route to ClickUp or Slack
New Intercom message? The agent classifies it by category and urgency.
This directory currently includes 15 recipes for the triage & route pattern, across 9 roles. Triage and routing agents turn an incoming item into a clear next step. They read the request, identify the relevant signal, and send it to the person, queue, or system that can act. This pattern is useful when a team receives more tickets, leads, applications, alerts, or messages than it can inspect consistently by hand. The examples here show the pattern across different roles, but the underlying design stays practical: define the categories, make the destination explicit, and keep an escalation path for ambiguous or sensitive work. A good triage agent does not pretend to solve every request. It shortens the time between arrival and ownership, while making the decision easy to review. Every example makes its inputs, output, and handoff concrete, so you can compare real workflow boundaries before adapting the pattern to your own tools and process.
New Intercom message? The agent classifies it by category and urgency.
Captures Jotform feedback, uses AI to analyze sentiment, sends thank-you emails for positive reviews.
New Zendesk refund ticket? It checks the WooCommerce order, alerts Slack for damaged items, or emails customers for proof on other cases.
Monitors Discord messages, uses AI to categorize requests, and routes them to the correct channels automatically.
Detects sensitive PII in chat inputs using risk scoring and routes them through secure local sanitization or cloud LLMs with full audit...
Captures Jotform feedback, uses AI to analyze sentiment, sends thank-you emails for positive reviews.
Receives chat queries via webhook, uses Gemini to extract details, validates the department against WHMCS.
New Zendesk refund ticket? It checks the WooCommerce order, alerts Slack for damaged items, or emails customers for proof on other cases.
Submits KYC details via form to AI for instant risk classification.
Routes incoming messages to specialized AI agents using Notion context for personalized support or automated booking.
New Jotform feedback is triaged instantly: urgent alerts for negative sentiment, auto-replies to questions via Gmail.
Automatically triages your Gmail inbox using AI to filter noise, summarize newsletters.
Receives support tickets via webhook, classifies issues with OpenAI, manages user records in Xano.
Automatically triages your Gmail inbox using AI to filter noise, summarize newsletters.
Monitors a Notion database for status changes and routes alerts to specific team members via Slack or other channels.
New Jotform feedback is triaged instantly: urgent alerts for negative sentiment, auto-replies to questions via Gmail.
Automatically triages your Gmail inbox using AI to filter noise, summarize newsletters.
Submits KYC details via form to AI for instant risk classification.
Monitors a Notion database for status changes and routes alerts to specific team members via Slack or other channels.
Detects sensitive PII in chat inputs using risk scoring and routes them through secure local sanitization or cloud LLMs with full audit...
Receives support tickets via webhook, classifies issues with OpenAI, manages user records in Xano.
Turns fitness logs into a pirate adventure where an AI scores your activity and routes you to specialized agents for advice.
Receives a WhatsApp message, looks up the sender's address in Google Sheets, and alerts the on-duty staff for that specific zone.
Routes incoming messages to specialized AI agents using Notion context for personalized support or automated booking.
New Intercom message? The agent classifies it by category and urgency.
New GoHighLevel contact? GPT-4 scores it 1-100, tags the lead, assigns a rep, and pings Slack for hot prospects.
It classifies an incoming item and sends it to the right queue, person, or next step based on rules and context.
It works best in high-volume workflows with repeatable categories, such as support tickets, leads, requests, and alerts.
Yes for unclear, sensitive, or high-impact cases. Clear routing rules and escalation paths make the pattern safer.