Write agent instructions and choose its actions

An agent does what its Instructions say, and nothing past replying unless an action under Actions is switched on. Both sit in the Configuration section of the agent editor, on the AI Agents tab of Inbox.

Write the instructions

To brief the agent:

1. Open Inbox, then AI Agents, and press an agent’s Edit.
2. Type into the Instructions box, in short headed sections.
3. Use Add prompt templates to drop a ready-written section in.
4. Press Optimize if you’d like it rewritten more clearly.

Under the box, Learn how to write this opens How to write instructions, which suggests five parts in order: who the agent is, what its first reply does, the steps of a good conversation, what it must never do, and then a test.

Add prompt templates offers Greeting, Qualifying questions, Booking a call, Follow-up on silence, Escalation and Boundaries. Whichever you pick lands at the end, after what you’ve already written.

Optimize keeps every fact, rule, name and number and rewrites the wording step by step. It refuses a rewrite that drops the customer’s name or comes back much shorter than yours, and leaves your text as it was. The message after a rewrite offers Undo.

Put a name in

The @ button under the box types the name of the advisor or the business this agent speaks for. The {} button inserts Customer's first name, which goes in as $contact.firstname.

Before each reply, the agent reads it as the first word of the customer’s name. A conversation with no real name, such as one labelled with a phone number, gets “there” instead, so the greeting reads “Hi there!”.

Choose what it may do

Each switch under Actions hands the agent one more thing it can do. Open the info mark beside an action’s name for a one-line description.

• Close conversations closes the thread with a short summary and whether it ended won, lost or neither. The summary goes into Private note under the thread’s Details.
• Assign to you or your team stops the agent and hands the conversation to a person, with a reason. You’re sent a WhatsApp at your Tell me on WhatsApp number.
• Update lifecycle stages moves the lead between new, replied, booked, won and lost.
• Update contact fields fills an empty field on the contact with a detail the customer wrote, such as a name or an email. Anything else waits under AI agent on the contact, for you to Use this or Dismiss. See Review what the AI agent heard.
• Update tags adds or removes tags on the contact.
• Add comments leaves notes for you in Private note, which the customer never sees.
• Follow up on inactive conversations checks in when the customer goes quiet.

Trigger workflows, Handle calls and Make HTTP requests show Not available yet and can’t be switched on.

Switch an action on and a box opens under it headed When and how should this action be performed?. Write when the agent should use it. Each of those boxes has its own Add prompt templates and Optimize, with templates written for that action.

Set the follow-up

Follow up on inactive conversations takes two numbers: after how many minutes of silence, and up to how many times. It starts at 5 minutes and 2 times. Minutes run from 1 to 1,440 and times from 1 to 5.

The agent is only woken when its own side spoke last, and the count starts over when the customer writes. What it says on each round comes from your instructions and the guideline box.

Set its limits

Press the gear button in the editor’s header to open the agent’s settings.

Stop after this many replies ends a long conversation and hands it to you. It’s 12 by default, and it takes 1 to 40. Quiet from and Quiet until are Singapore time, and a message that lands between them gets no answer. Languages names the languages it handles, and it’s told to answer in the one the customer writes in.

A district admin also sees Model and Effort there.

Rules every agent follows

Every agent is also given rules after yours, and your instructions can add to them but not remove them. It’s told never to give financial, investment, insurance, tax or legal advice, never to quote prices, premiums or returns, and never to recommend a specific product, fund or provider.

It’s also told never to claim to be a person, and never to ask for an NRIC, bank details, card numbers, passwords or documents. A complaint, a mention of a regulator or PDPA, or a question its knowledge can’t answer should end with the conversation handed to you.

Note

A customer who asks not to be contacted is always stopped, whichever actions are switched on. The conversation is closed and marked lost.

FAQ

Is there a limit to how long the instructions can be?

Yes. Anything past 40,000 characters is cut when it saves, and each action’s guideline keeps its first 4,000. The count under the box shows how many you’ve used.

What happens to a detail my book has no column for?

It waits on the contact, under AI agent and Not recorded. A button such as Add Occupation field makes the column and fills it in.

Will the agent change a value I typed?

No. It fills empty fields, or corrects a value it wrote itself, and only with the customer’s own words. A different value waits under AI agent on the contact. Type the same value in yourself and the suggestion clears.

Do changes go live as I type?

No. Nothing is saved until you press Save draft or Publish. On a published agent the button reads Publish changes, and that save is live straight away.

Does the template already switch actions on?

Yes. Receptionist starts with assigning, following up and closing. Sales agent has those three plus stages, contact fields and comments, while Support agent has the three plus comments.