Classify Intercom Messages & Route to ClickUp or Slack
New Intercom message? The agent classifies it by category and urgency.
What it does
You get incoming Intercom messages automatically sorted by category and urgency so you can focus on the right work. The agent creates structured tasks in ClickUp for operational requests or sends real-time alerts to Slack when a sales lead arrives.
Step by step
- Configure Intercom webhooks to POST JSON payloads containing conversation metadata and message bodies to your server endpoint.
- Define a strict JSON schema for the OpenAI response that includes fields for category, urgency score, and routing destination.
- Implement conditional logic in your code to check the 'destination' field from the AI output before triggering external API calls.
Where the LLM does the work
- Prompt engineering is required to create a system instruction that accurately distinguishes between vague product feedback and urgent sales inquiries.
- The LLM must be instructed to extract specific action items or context snippets to populate the ClickUp task description effectively.
- You need the model to assess sentiment nuances so it can correctly flag high-priority issues for immediate Slack alerts.
Watch out for
ClickUp task creation might fail if required custom fields are missing in the destination list, validate field IDs during setup.
Slack message formatting can break if the AI output contains unescaped special characters, sanitize all text before sending.
Tools that fit
The agent brief
Everything your agent needs, including the gotchas. Copy it and go.
You are helping me build the following AI agent workflow. ## Goal Classify Intercom Messages & Route to ClickUp or Slack: New Intercom message? The agent classifies it by category and urgency. ## Specification - What it does: You get incoming Intercom messages automatically sorted by category and urgency so you can focus on the right work. The agent creates structured tasks in ClickUp for operational requests or sends real-time alerts to Slack when a sales lead arrives. - Trigger: Triggered by an event (Event · on new conversation) - Autonomy: Fully hands-off - Expected setup effort: about an afternoon - Tools/services involved: - Intercom: Webhook intake of new conversations - OpenAI Chat Model: Classify category, sentiment, and urgency - Code: Parse AI output and merge with details - ClickUp: Create tasks for Support or Product requests - Slack: Send alerts for Sales inquiries ## Known pitfalls, handle each one explicitly in your implementation 1. ClickUp task creation might fail if required custom fields are missing in the destination list, validate field IDs during setup. 2. Slack message formatting can break if the AI output contains unescaped special characters, sanitize all text before sending. ## Reference implementation https://n8n.io/workflows/7933 (workflow template) Fetch and inspect this before building. If it matches my stack, adapt it; if not, rebuild the pattern with my tools. ## Process requirements 1. Before building: ask me which of the listed tools I actually use and what my platform is (n8n / Make / code / other). Do not assume. 2. Adapt the pattern to my answers; do not force the reference stack. 3. Address every pitfall above; tell me how you handled each. 4. Provide a test plan I can run before letting this touch real data. 5. Ask before any step that sends messages, modifies data, or spends money. Source: https://usecasesforagents.com/use-case/classify-intercom-messages-route-to-clickup-or-slack/ via usecasesforagents.com
Frequently asked questions
Can I use a different Support & Ticketing than Intercom?
Yes. Intercom is only the example support & ticketing in this recipe. The same flow works with Hiver, Marker.io, Syncro and WHMCS API. Swap the support & ticketing connection and keep the rest of the setup as written.
Want this running in your business?
This is what I do. I design and build AI agents like this one, and keep them running. If you want it set up for your team instead of doing it yourself, get in touch.
Who it's for
Built for support teams drowning in inbound and looking to triage faster, not just harder.
Sales teams can plug this into their own CRM and call cadence just as easily.