Zendesk's Ticketing API lets an agent read tickets, set fields and add comments, and a comment can be private. We build the agent so its replies land as internal notes, and a person on your support team decides what the customer sees.
01Use
Each one does a single job on the Ticketing or Help Center API. Replies arrive as internal notes, and nothing reaches the customer without a person.
Reads each new ticket and the requester's history, sets type, priority and tags, and routes it to the right group.
Drafts a reply from your help center and past tickets and adds it to the ticket as an internal note for the agent handling it.
Finds questions your help center does not answer and drafts article outlines for your team to review.
02Wire
Five surfaces, all public Zendesk APIs and frameworks. The agent works as its own API user, so its actions show up in ticket history.
Read access is broad enough to answer the question, write access is narrow, and every write has a gate until you decide otherwise.
| Agent | Reads | Writes | Human gate |
|---|---|---|---|
| Ticket triage | tickets, users |
Fields, tags, group | Sensitive priority changes |
| Drafted replies | Ticket history, articles | Internal note | Every public reply |
| Help center gaps | Articles, tickets | Drafted outlines | Every article |
03Gate
Zendesk already separates internal notes from public replies. The agent writes notes; your support agents decide what the customer sees.
On the ticket itself. The draft is an internal note, and the sidebar app shows where the answer came from.
The drafted reply, the help center articles and past tickets it drew on, and the run trace behind it.
Every public reply, every published article, and priority changes on tickets you mark as sensitive.
An edited or discarded draft is recorded with its trace and becomes an eval case, so the next draft is tested against it.
04Asked
No. The agent writes internal notes. A person on your team turns a draft into a public reply, unless you tell us in writing to change that for a specific type of ticket.
No. It works alongside them. Triggers can start the agent, and the agent can suggest the macro your team would normally apply.
From your help center and your past tickets, retrieved for each ticket and cited in the draft, so the person reviewing it can check the source.
Tell us the question your support team answers most. We will tell you what the agent would read, what it would draft, and how your team would review it.
