Configure prompts for Breeze agents with HubSpot MCP Client

After connecting your Breeze agents to an MCP Server using HubSpot MCP Client, you must add clear prompt instructions to guide your Breeze agent on which MCP tools to use and when. The prompts should be customized based on your specific use case, setup, and goals. 

Add MCP prompts to Breeze agents

1. In your HubSpot account, navigate to Breeze > Breeze Studio.
2. Navigate to the Agents tab, then click the name of the agent you want to add MCP prompts to. 
3. In the top right, click Configure.
4. In the Guide how this agent behaves section, in the Extra instructions text field, enter your MCP prompts.
   
   Example prompt: 
   You are connected to the G2 MCP Server. For every deal analysis, always query the G2 MCP Server to retrieve buyer intent signals, competitor comparison activity, and intent scores for the companies associated with closed-lost deals. Combine G2 external market signals with HubSpot CRM data to provide a complete picture of loss reasons. When a competitor is identified in G2 data, cross-reference it against the deal's loss reason in HubSpot. Surface patterns where G2 intent activity spiked or shifted toward a competitor before deal close.

Best practices for adding MCP prompts to Breeze agents

Use the following examples as a guide for adding MCP prompts to a Breeze agent connected to a Notion MCP Server. While these best practices offer a general framework, you can customize them for your specific use case and for other MCP Servers.

Identify the database

Since Notion pages and databases don't follow a fixed structure, you need to instruct the agent exactly on which database to use. It’s recommended that you mention the name of the database, its purpose, and include the database ID if you have it.

Example prompt:

Use the Notion database titled 'Content Projects Tracker,' which includes the project name, status, owner, and due date.

Describe the property or column structure

Describe your database structure to help the Breeze agent understand the data. It’s also recommended that you list the relevant column names, what they contain, and their property type.

Example prompt: 

The database includes these columns: 'Project Name' (title), 'Owner' (person), 'Status' (select), and 'Deadline' (date).

If some fields are optional or not consistently used, it's helpful to mention that as well.

Define the agent’s task

Be clear about the specific information the agent should find or extract.

Example prompts:

• Find all items where the Status is 'In Progress' and the Deadline is within the next 7 days.

• Summarize all entries that have 'Product Launch' in the title.

Specify actions and tools

If your agent needs to perform a specific action, such as query_database, list_databases, or retrieve_page, it’s helpful to explicitly state which tools it should use and how. This is especially useful if the agent has to choose between multiple tools.

Provide an example input and output

If your agent needs to add or edit entries, it’s recommended that you include an example of what the data should look like.

Example prompt:

When creating a new entry, fill in 'Status' as 'Not Started' by default and leave 'Deadline' blank.<

Examples of MCP prompts for Breeze agents

You are connected to the Notion MCP server. Use the database titled 'Feature Rollouts Tracker'. It includes the following columns: 'Feature Name' (title), 'Status' (select: Not Started, In Progress, Done), 'PM Owner' (person), and 'Target Release Date' (date). Your job is to find all features that are marked 'In Progress' and return their names along with the PM Owner. Use the query_database tool and only pull results from this database.

Analyze all closed-lost deals from the last 90 days. For each company, check G2 intent signals to see if they were actively researching the competitor listed in our loss reason. Do the G2 signals confirm or contradict what our reps reported?

Which competitors are appearing most frequently in both our HubSpot closed-lost reasons and in G2 comparison activity for those same accounts? Rank them and tell me where our reps' notes align or conflict with what buyers were actually doing on G2.

Analyze our closed-lost deal patterns from the last 90 days using G2 intent data. For each major loss theme -- pricing, competitor preference, no decision -- create an Asana task in our Sales Enablement project with a recommended action, owner placeholder, and due date 2 weeks from today.

For each company in my 'High Priority Targets' list, run a company research profile and layer in their current G2 buyer intent data. Flag any account with an intent score above 65 as hot and update their CRM record accordingly.

Pull company research on our 5 most recently created leads and cross-reference each with G2 intent data. Tell me which ones are already in an active buying cycle based on G2 signals and what they're looking at.

Based on G2 intent data, which competitor are we most commonly being compared against this month? Write a blog post optimized for that head-to-head comparison that a buyer would land on mid-evaluation.