The Problem
Routine support questions still require care. A useful reply may depend on the exact policy, the customer's wording, and details that are not present in the first message. A fast but invented answer can be more damaging than a slower response that clearly asks for missing information.
The practical first step is supervised drafting: help a person produce a consistent response while keeping account checks, exceptions, refunds, and final delivery outside the Ant.
What You Can Use Today
Support Ant is an available preset for the local request desk. It drafts only from information a person deliberately supplies and cannot inspect a customer account, retrieve an order, change a subscription, issue a refund, close a case, or send a message.
1. Create a supervised Support Ant
Choose the Support Reply preset in My Ants, then tailor its tone, purpose, instructions, signature, and small trusted knowledge notes. Human review remains locked.
2. Add the customer request manually
Paste the request into the local request desk and assign the Support Ant. Include only context you are permitted to use; ANTS has no help-desk or account connection.
3. Prepare and review a reply
Write the draft yourself or explicitly confirm optional AI drafting. Edit every factual statement, make missing context visible, and approve the exact revision you reviewed. Copying the result is a manual handoff, not proof that the customer received it.
What Remains on the Roadmap
A connected Support Ant could eventually read an authorized queue, classify new cases, retrieve permitted customer context, and propose routing or resolution steps. That would require accounts, scoped integrations, durable audit history, permissions, error handling, and explicit approval gates. None of those connected actions are part of the current local product.
- Available now: local profile configuration, manual request intake, editable reply drafts, immutable approval revisions, and manual copy handoff.
- Not available now: ticket ingestion, customer lookup, sentiment monitoring, automatic escalation, case updates, refunds, or sending.
- Safe pilot: use low-risk questions with known policy context and record where the draft needed factual correction or escalation.