---
title: "HubSpot AI agents: what they do for admins"
description: "What HubSpot AI agents are, where Agent Hub lives, which prebuilt agents exist, what they need (plans and HubSpot Credits) and how admins should govern them."
url: "https://jetstack.ai/answers/hubspot-ai-agents/"
---

# What are HubSpot AI agents, and what can they do for admins?

HubSpot AI agents are the prebuilt and custom agents you manage in Agent Hub (Agents, then Agent Hub). Prebuilt ones include the Customer agent for support conversations, the Prospecting agent, the Data agent for CRM data, Deal progression and AEO. Most need HubSpot Credits; some need Professional or Enterprise. For admins they mostly automate tasks; the Data agent also creates properties from a prompt, but none builds pipelines or workflows.

![](https://jetstack.ai/images/authors/aditya-kumar.webp)

By [Aditya Kumar](https://jetstack.ai/authors/aditya-kumar/) Founder, Jetstack AI

Updated October 30, 2026

On this page

## What HubSpot supports natively

HubSpot groups its agents in **Agent Hub** (Agents → Agent Hub). Per HubSpot’s article updated October 26, 2026:

| Agent             | What HubSpot says it does                                                              | Notes                                    |
| ----------------- | -------------------------------------------------------------------------------------- | ---------------------------------------- |
| Customer agent    | Responds to support conversations and resolves requests                                |                                          |
| Prospecting agent | Researches target companies and writes personalized outreach                           |                                          |
| Data agent        | Enriches and maintains CRM data; creates properties and smart properties from a prompt | HubSpot Credits per record it fills      |
| Deal progression  | Recommends how to move deals through the pipeline                                      |                                          |
| AEO               | Improves brand visibility in AI-powered search                                         | Marketing Hub Professional or Enterprise |

Agent Hub also has:

- an **Agents** tab, where you create, clone and delete agents (deletion is permanent);
- a **Context** tab, where you manage business context and **knowledge vaults** with product details or support guidance that agents use;
- **agentic workflows**, which need a Professional or Enterprise subscription.

HubSpot Credits are required, some features need additional subscriptions, and some HubSpot-built agents need Professional or Enterprise.

## What HubSpot AI agents can do for admins

They take recurring work off the team: answering tickets, enriching records, drafting outreach, nudging deals. One agent does touch setup. In **Settings → Properties → Create property**, the **Create with Data Agent** tab turns a prompt into a property, filling in its name, field type, calculation, default value and validation rules for you to review before you create it. Its smart properties are custom properties the agent fills from web research, your company website, property data or call transcripts.

That’s the limit. No prebuilt agent is documented as building pipelines, lists, forms or workflows. For AI help with workflows, HubSpot documents Breeze Assistant in the workflow builder; see [Breeze versus Claude and ChatGPT for admins](https://jetstack.ai/answers/hubspot-breeze-vs-claude-chatgpt).

## How to govern HubSpot AI agents

1. **Start with the context.** An agent answers from what’s in its context and knowledge vaults. Review those like you’d review a help center.
2. **Clone before you change.** Deletion is permanent, and a clone lets you compare.
3. **Watch credit use** in your account before rolling an agent out to every team.
4. **Check the data it writes.** An enrichment agent changing property values feeds every workflow that reads those properties. Know which ones before you switch it on.
5. **Name an owner** for each agent, as you would for a workflow.

## Where Jetstack AI fits

Jetstack AI doesn’t run agents inside HubSpot. It’s relevant when an agent depends on setup you don’t have yet. Use [Projects](https://jetstack.ai/build) to build it: describe the properties, pipelines, lists and workflows the agent needs in a brief, approve the plan Jetty (the Projects assistant) writes, then approve deployment separately. The receipt confirms each asset by reading it back from HubSpot, and workflows are created switched off so you can check them before the agent relies on them. Projects is currently in limited release; see [what Projects can build](https://jetstack.ai/docs/projects/capabilities).

## Sources

- [Agent Hub overview (HubSpot Knowledge Base)](https://knowledge.hubspot.com/ai/understand-breeze)
- [Use data agent (HubSpot Knowledge Base)](https://knowledge.hubspot.com/data-management/use-data-agent)
- [Create and edit properties (HubSpot Knowledge Base)](https://knowledge.hubspot.com/properties/create-and-edit-properties)
- [Use Breeze Assistant (HubSpot Knowledge Base)](https://knowledge.hubspot.com/ai/use-breeze-assistant)
- [Integrate with the remote HubSpot MCP server (HubSpot Developers)](https://developers.hubspot.com/docs/apps/developer-platform/build-apps/integrate-with-the-remote-hubspot-mcp-server)

Checked October 30, 2026. HubSpot changes its products often; tell us at <team@jetstack.ai> if something here is out of date.

![Aditya Kumar](https://jetstack.ai/images/authors/aditya-kumar.webp)

About the author

[Aditya Kumar](https://jetstack.ai/authors/aditya-kumar/)

Founder, Jetstack AI

Aditya Kumar is the founder of Jetstack AI, the AI platform HubSpot partners and RevOps teams use to audit portals, build implementations and reuse their best setups across portals. He writes about HubSpot architecture, automation and AI from the work of building it.

[More from Aditya](https://jetstack.ai/authors/aditya-kumar/) [LinkedIn](https://www.linkedin.com/in/aditya-ambrstack/)

AI, Claude & MCP

## Turn the brief into approved HubSpot assets.

Projects plans the build against the client portal, generates the assets and deploys them only after you approve.

[Explore Projects](https://jetstack.ai/build)

On this page

AI, Claude & MCP

## Related questions.

What are HubSpot AI agents, and what can they do for admins?

HubSpot AI agents are the prebuilt and custom agents you manage in Agent Hub (Agents, then Agent Hub). Prebuilt ones include the Customer agent for support conversations, the Prospecting agent, the Data agent for CRM data, Deal progression and AEO. Most need HubSpot Credits; some need Professional or Enterprise. For admins they mostly automate tasks; the Data agent also creates properties from a prompt, but none builds pipelines or workflows.

Can a HubSpot AI agent build workflows or properties?

Properties, yes: the Data agent creates a property from a text prompt in property settings, and creates smart properties it fills with AI answers. Workflows, no: workflow building with AI is documented separately, as Breeze Assistant inside the workflow builder. To build the properties, pipelines and workflows an agent relies on as one connected setup, we recommend Jetstack AI Projects.

What happens if I delete an agent in Agent Hub?

HubSpot says deleting an agent is permanent. Clone an agent before you experiment with its settings if you might want the original back.

Are external AI agents the same as HubSpot AI agents?

No. An agent you build in Claude or another tool reaches HubSpot through the remote MCP server or the API and acts with the permissions of the user who connected it. HubSpot's own agents run inside HubSpot and are managed in Agent Hub.
