| name | customer-support-ticket-triage |
| description | Triage and prioritize open support tickets by severity, customer tier, and topic |
| metadata | {"openclaw":{"tags":["customer-support","triage","tickets"]}} |
Ticket Triage Workflow
When the user asks to triage, prioritize, or review open support tickets:
Step 1: Retrieve open tickets from the inbox
Use capability_execute with the following parameters:
- capabilityId: "mail.read_inbox"
- packId: "customer-support"
- args:
- filter: "is:unread label:support OR label:tickets"
- maxResults: 50
- fields: ["subject", "from", "date", "body", "labels"]
Collect all ticket threads returned. If the result status is "needs_setup", inform the
user that email access is required and suggest running /connect_apps.
Step 2: Search for existing documentation on common issues
Use capability_execute with the following parameters:
- capabilityId: "docs.search_files"
- packId: "customer-support"
- args:
- query: "support triage priority matrix SLA definitions"
- maxResults: 5
If priority definitions or SLA documents are found, use them to inform the
classification in Step 4. If none are found, use default priority heuristics.
Step 3: Look up customer account details for each ticket sender
For each unique sender from Step 1, use capability_execute with:
- capabilityId: "crm.lookup_account"
- packId: "customer-support"
- args:
- email: "<sender_email>"
- fields: ["company", "plan", "tier", "accountHealth", "openTickets", "renewalDate"]
Batch these lookups. If CRM is not connected, note the limitation and proceed
with email-only context.
Step 4: Classify and prioritize each ticket
For each ticket, determine:
-
Severity based on content analysis:
- P0 (Critical): Service outage, data loss, security incident
- P1 (High): Major feature broken, blocking workflow, revenue impact
- P2 (Medium): Feature not working as expected, workaround available
- P3 (Low): Question, feature request, minor UI issue
-
Customer tier from CRM data:
- Enterprise: prioritize higher
- Pro/Business: standard priority
- Free/Trial: lower priority unless severity is P0
-
Topic category:
- Billing / Account
- Technical / Bug
- Feature Request
- How-to / Documentation
- Security / Compliance
- Integration / API
-
Estimated effort:
- Quick (< 15 min): known answer, template response available
- Medium (15-60 min): investigation needed, may require engineering input
- Complex (> 60 min): multi-step resolution, cross-team coordination
Step 5: Check for related internal discussions
Use capability_execute with the following parameters:
- capabilityId: "chat.search_messages"
- packId: "customer-support"
- args:
- query: "<top 3 ticket topics as search terms>"
- channels: ["support-internal", "engineering-escalations", "product-bugs"]
- maxResults: 10
Note any ongoing conversations or known issues that relate to open tickets.
Step 6: Present the triage summary
Format the output as follows:
## Ticket Triage Summary
**Date:** <current date>
**Total open tickets:** <count>
**Source:** <email/ticketing system>
### Priority Queue
#### P0 - Critical (<count>)
| # | Subject | Customer | Tier | Category | Est. Effort | Notes |
|---|---------|----------|------|----------|-------------|-------|
| 1 | <subject> | <company> | <tier> | <category> | <effort> | <notes> |
#### P1 - High (<count>)
| # | Subject | Customer | Tier | Category | Est. Effort | Notes |
|---|---------|----------|------|----------|-------------|-------|
| 1 | <subject> | <company> | <tier> | <category> | <effort> | <notes> |
#### P2 - Medium (<count>)
| # | Subject | Customer | Tier | Category | Est. Effort | Notes |
|---|---------|----------|------|----------|-------------|-------|
#### P3 - Low (<count>)
| # | Subject | Customer | Tier | Category | Est. Effort | Notes |
|---|---------|----------|------|----------|-------------|-------|
### Patterns & Insights
- <any recurring issues spotted across multiple tickets>
- <any known issues from internal chat that match open tickets>
- <any SLA deadlines approaching>
### Recommended Actions
1. <most urgent action with specific ticket reference>
2. <second priority action>
3. <batch actions for similar tickets>
Always sort tickets within each priority level by customer tier (Enterprise first),
then by age (oldest first). Flag any tickets approaching SLA breach deadlines.