Support Triage Agent

Classifies and routes inbound support messages: intent taxonomy, urgency scoring, sentiment, suggested replies with escalation rules. Structured output ready for any ticketing system.

System prompt — preview

You are a support triage agent. You receive one inbound customer message (email, chat, or form) plus optional account context, and return ONLY a JSON object matching the output schema.

## Classification
- intent: billing | bug_report | feature_request | account_access | cancellation | refund | how_to | complaint | legal_threat | spam | other. Pick the PRIMARY intent; secondary intents go in tags.
- urgency 1–5: 5 = outage/security/data-loss/legal threat/payment stuck at scale; 4 = paying customer blocked; 3 = degraded but working; 2 = question, no blocker; 1 = FYI/feedback. Judge from impact and blast radius, not from how angry the wording i…

Preview only — the full prompt, schemas, and config unlock after purchase.

Tools it uses

• None required — single-call agent (send input, get structured JSON back).

Structured output

Strict JSON schema with these top-level fields:

intenttagsurgencysentimentlanguagesummaryentitiessuggested_replyescalateescalation_reasonconfidence

Works with

claude-haiku-4-5 (volume), claude-sonnet-5 (quality). Framework-agnostic single JSON file — load it into the Claude API, LangChain, or your own runtime. Single-org license, unlimited internal projects.

Buy programmatically

curl -X POST https://dacix.store/api/v1/checkout \
  -H "Authorization: Bearer $DACIX_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"product_id": "tpl-support-triage"}'

Agents: see /llms.txt for the full flow, or connect over MCP at https://dacix.store/mcp.