Offload routine balance-check calls to a verified voice agent that understands account nicknames, watches caller sentiment, and hands stressed customers to a person with full context: AI agent prompt
Salesforce's customer story on Southern Cross Credit Union (SCCU) in Australia describes Nova, an Agentforce Voice agent built with Forward Deployed Engineers in six weeks and integrated with a Vonage contact center IVR. Callers choosing the balance option are verified, then Nova reads balances using natural language—including product codes and nicknames instead of formal account numbers—and escalates on frustration or vulnerability criteria with full context. SCCU reports Nova providing balances to 95% of verified callers, handling 7% of all customer calls in an early month despite only covering balances, and about a 4% reduction in total staff hours on that volume. Architecture keeps the LLM from directly accessing customer data: Nova collects inputs and uses deterministic backend tools via Service Cloud. Wait times had risen from under two minutes toward eight before the launch. Figures are SCCU's as reported by Salesforce.
The prompt
Design a verified balance-check voice agent for my contact center. Stack: [CCaaS/IVR], core banking: [system], verification: [method stronger than OTP if possible]. Rules: never speak a balance before verification succeeds; never let the LLM read raw customer records directly; escalate fraud, scams, distressed sentiment, and vulnerability flags immediately. 1) Write the IVR → verify → balance → wrap-up flow. 2) List nickname/product-code examples the agent must understand. 3) Draft sentiment and vulnerability handoff rules. 4) Propose CRM Analytics metrics: connect rate, verify rate, balance-success rate, pre/post escalation. 5) Give a six-week build-and-test plan including accent simulations. Do not look up a real account in this chat.
Fill in the blanks, then copy
This prompt has 3 [bracket]s to fill in. Type your details below and the prompt updates as you go. Your answers stay in this browser and are reused on other prompts with the same blanks. Nothing is sent to us.
Example values are fictional, just to show how a filled-in prompt reads. Replace them with your real details.How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (use the fill-in-the-blanks box above, which remembers your answers for the next prompt).
- Paste it into Salesforce Agentforce Voice, Service Cloud, Vonage (or your CCaaS) and review the answer before you use it.
Works with: Salesforce Agentforce Voice, Service Cloud, Vonage (or your CCaaS), ChatGPT, Claude, MuleSoft/core banking APIs
Real-world example: salesforce.com/au/customer-stories/southern-cross-credit-union-sccu/
Featured in: AI prompts for writing SOPs, reports, policies and checklists
Before you use it: what to check
Already built into this prompt
- Rules: never speak a balance before verification succeeds; never let the LLM read raw customer records directly; escalate fraud, scams, distressed sentiment, and vulnerability flags immediately.
- Do not look up a real account in this chat.
Check yourself
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically".
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it freeMore Customer Support prompts
See all Customer Support prompts →