Customer-support agent with Claude tool-use

Multi-tool agent: search KB, escalate to human, refund processor. Tools defined in TypeScript, Claude orchestrates via function-calling.

70% containment rate on first 1000 conversations.

0

Problem

Support team of 4 was drowning in tier-1 tickets — KB lookups, order status, simple refunds — leaving no bandwidth for hard issues.

Users

Public — embedded in the customer help widget, ~400 conversations/day

Key flow

1Customer types question in widget
2Agent picks tool(s), confirms intent if ambiguous, acts
3Resolution or handoff — transcript logged for QA review

What I built

Claude function-calling orchestrator with 6 typed TypeScript tools
Vector KB search over docs + past resolved tickets
Refund tool gated by order-age and amount thresholds
Live handoff to human agent when confidence drops below 0.6

Limits

No payment-method changes — always escalates to a human
Knowledge cutoff: KB index is rebuilt nightly, not real-time

Customer-support agent with Claude tool-use

AI tools used

Claude
Anthropic API
Supabase

@lukas.novak