NetApp Iterate.ai
NetApp Sellers & Partners
Joint Educational Series
Field Brief · The Building Blocks

How an Agent Actually Works

Every brief in this series uses the terms “agent” and “tool” as if everyone knew what that means. This brief explains those terms.

What you'll learn
What actually happens between your question and its answer
What a “tool” is, and why a model needs one at all
The one standard that lets an agent use almost any tool

The core argument

A chatbot is a simple conversation. You ask, it answers from what it already knows. An agent has a job list: It can stop mid-answer, go get something it does not know, and come back — then decide if it needs to go get something else before it is done.

Every one of those trips is a tool call. An AI model can write and reason, but it cannot check today's date, look up a record, or send an email by itself. The model needs tools to do those tasks. A tool extends the AI with one specific, narrow action it can trigger.

A tool can be almost anything a business already uses. If a person could look something up or take a specific action, an agent leverage a tool that does the same thing — including searching the open web.

Email (e.g. Gmail)CalendarSpreadsheet (e.g. Excel) Accounting (e.g. QuickBooks)A database (e.g. Snowflake)The open web Messaging (e.g. Slack)CRMSupport ticketsA calculator
One question, from prompt to answer
1
You ask
This is the prompt
2
Agent decides
Do I need a tool for this?
3
Tool runs
Looks something up or perfoms a task
4
Answer
Comes back after potentially several trips

Steps 2 and 3 repeat as many times as the job needs — the agent decides, one trip at a time.

A worked example

“What is my copay for tomorrow's visit?” The agent makes two tool calls to find out.

Prompt
Look up insurance tier
Look up appointment type
“$40”

Key facts

A prompt is the instruction. A tool is the action. The agent decides when to use which.
Most of what an agent does is invisible to the person who asked the question.
A tool can be as simple as a calculator or as complex as a full database query.
Every tool call is an instance for data to stay secure or leak out of your walls.
NetApp | Iterate.aiHow an Agent Actually Works · 01 / 02

The standard that lets it plug in anywhere

Without a shared standard, every agent would need custom wiring for every tool it might use — a separate connector per pair, rebuilt each time either the agent or tool changes.

MCP — Model Context Protocol — fixes that. It is a shared job description any tool can implement. Once a tool understands this protocol, any agent that also speaks it can find that tool, see what it does, and use it. Think of it like a wall socket instead of a bare wire: once a machine has the right plug, it works in any socket built to the same standard.

MCP is not one company's product — it is an open specification. Database providers, SaaS applications, and common data sources are now providing their services and data as an MCP tool, so any agent can query or interact without a custom connector.

41%
Of software organizations already run MCP-connected tools in production
10,000+
Active public MCP tools an agent could potentially reach
970×
Growth in monthly adoption of the standard in its first eighteen months
Why this is a governance question too

Every tool an agent can call is also a door. Standards that makes doors easy to add will also make it easy to add too many. An unreviewed tool is the same risk as an unapproved app, but it's faster, and running all day.

Key facts

MCP decides how a tool is described and called, but not what it is allowed to do. Permissions are a separate decision controlled by the overall agent platform.
A tool that can talk to an MCP can also be found and used by any agent that also speaks that same protocol with no custom code.
The same security rule applies here as to every model call: know what it is, and log what it did.
Further reading

Full paper, 10 pages — iterate.ai/partners/netapp/private-ai-education-and-certification-program. On deciding which model handles a job, see What Is a Harness?; on seeing every call after the fact, You Cannot Govern What You Cannot See.

About this series

A joint educational series on private AI and the AIPod Mini. NetApp® — the governed data-control layer. Iterate.ai — the private intelligence layer.

NetApp AIPod Mini
NetAppIterate.ai
v1.7 · Nov 2026
NetApp | Iterate.aiHow an Agent Actually Works · 02 / 02
Prev
No. 2 of 7
Next