# AI support agent that resolves, not deflects

A support agent answers from your knowledge base in your tone, follows your policies, and takes actions (refunds, account changes) through APIs, handing over to people when unsure.

**Why it matters:** Klarna and Intercom's customers show the ceiling: most routine questions can be resolved end to end, and human agents move to harder work.

- Trigger: Customer message (chat / email)
- Tools: Intercom Fin, Claude, OpenAI, Zendesk, Stripe
- Time saved (rough estimate): 20–60 hours/week
- Typical agency build cost (rough estimate): $5k–$40k

## Steps
1. Retrieve from help center + policies
2. Answer or plan an action
3. Call API: refund, change plan, look up order
4. Unsure or sensitive → human
5. Resolution + CSAT tracking

## Pitfalls
- Evaluate against real past conversations before launch, and on every change.
- Actions need strict limits (amount caps, allowed operations) and audit logs.
- Always offer a human; Klarna kept that option.

## Sources
- [Anthropic customer story: Intercom Fin](https://www.anthropic.com/customers/intercom)
- [Klarna press release: AI assistant handles two-thirds of chats](https://www.klarna.com/international/press/klarna-ai-assistant-handles-two-thirds-of-customer-service-chats-in-its-first-month/)

---
Source: https://cofounderbase.com/workflows/ai-support-agent-with-actions/?utm_source=md&utm_medium=ai · Cofounderbase, a directory of agencies that build AI internal tools and automations. Data comes from each agency's own website and official partner directories; rankings are not sold.
