Reference

Agent definitions

An agent's definition is its job description: the instructions, which AI model it uses, how long one turn may take, and which tools and skills it has.

The agent is the worker. The definition is how that worker currently does the job. Keeping those separate is what lets Scout get steadily better over months without ever losing its history.

The parts

PartThe question it answers
InstructionsWhat is this for, and what should it never do?
Model and reasoning levelHow much thinking does the work need?
TimeoutHow long may one turn run before we stop it?
ToolkitsWhich actions may it use?
SkillsWhich reusable instructions can it pull in for particular tasks?

Writing the instructions

This is the part that decides whether the agent is any good, and it's the part people rush.

Write it like you're briefing someone on their first morning. Four things:

What it's for. One sentence. "You sort incoming tickets for the Northwind IT service desk."

What good looks like. Be specific about shape. "Reply with a category, an urgency from 1 to 4, and a two-sentence summary. No preamble."

What to never do. This is the one people skip and then regret. "Never promise a client a timeline. Never say a problem is fixed unless you've confirmed it. Never guess at a licence count."

What to do when unsure. "If the ticket is ambiguous, say so and mark it for a human rather than guessing."

Keep the standing instructions about the enduring role. Put per-task procedures in skills and provider capabilities in toolkits, so the instructions stay short enough that you'll actually maintain them.

Picking a model

The model is the AI itself: Claude, GPT, Gemini, and so on. Different ones cost different amounts and are better at different things.

You don't need to have an opinion on day one. Start with the default, get the instructions right, and only then try a different model on the same task and compare. Changing the model and the instructions at the same time tells you nothing about which change helped.

Higher reasoning levels think harder and cost more. Worth it for judgment calls; wasted on "reformat this list".

The agent stays; the definition changes

You'll rewrite instructions many times. That doesn't create a new agent, and it doesn't throw away sessions, memory, or history.

SituationWhat to do
It's not quite doing the job wellChange the definition
You want it on a different modelChange the definition
It needs one more toolChange the definition
A different team should use it, with different accessNew agent
It's for a different clientNew agent, probably in a different project
It shouldn't be able to see what this one seesNew agent

Tuning it

Change one thing at a time. This is the entire method and it's easy to skip.

  1. Give it a real task in Chat.
  2. Read what it actually did, including which tools it reached for.
  3. Change one thing: a sentence of instruction, one tool, the model.
  4. Run the same task again and compare.

The session is a full record of what the agent saw and did, which is what makes this comparison possible at all. Change three things at once and you've learned nothing.

When your own software owns the behavior

Most teams write the definition in Mobius and edit it there. That's the right default and it keeps the behavior where operators can see it.

If you're building Mobius into your own product, you might already hold the per-customer behavior in your own system. There are three ways the definition can arrive:

SourceUse it when
Stored in MobiusYour team writes and operates the agent in Mobius
Sent by the callerYour software starts the conversation and already knows the customer's settings
Fetched from your backendMobius starts scheduled or event-driven work, but your backend owns the current behavior

These aren't different kinds of agent. They're different ways of supplying the same job description.

Reach for an application-owned definition when copying the same behavior into many client projects would drift out of sync. Otherwise store it in Mobius.

A definition your software sends can replace the stored tools and skills for that conversation, but it can never grant access the agent doesn't have. The Agent definitions API guide covers the exact rules, caching, and what happens when your backend doesn't answer.

In the app

Open Build > Agents, pick an agent, then Configure.

Start with clear instructions and no tools. Have one real conversation. Add the first tool only when something the agent genuinely needed was missing.

Next