How to Connect Breathe HR to Claude
Breathe HR holds your people data: employee records, departments, holiday and absence, leave requests, sickness, training, expenses and salaries. Once Breathe is connected to Claude, you can ask questions about all of it in natural language, such as who's off over Christmas or which department has the most sickness this year. Tugger syncs your Breathe data into a managed data warehouse and gives Claude read-only access to it.
What you can ask once Breathe HR is connected to Claude
Tugger syncs your Breathe data on a schedule into a warehouse that's yours, and Claude queries it through Tugger's read-only MCP server. Claude never writes back to Breathe.
Typical questions:
- Who's off in the weeks before Christmas, by department, and which requests are still waiting for approval?
- Which departments had the most sickness absence days in the last 12 months?
- How much holiday allowance does each department have left this year?
- Which training courses expire in the next 90 days?
- How many starters and leavers did we have each month this year?
Claude writes the query, runs it, and explains the result. You can follow up in the same conversation, ask for a chart, or ask Claude to build a dashboard.
What you need
- A Tugger account. You can start a free 10-day trial if you don't have one.
- Admin access to Breathe HR, so you can see the API settings.
- A Claude plan that supports connectors (Pro, Max, Team or Enterprise).
[VIDEO: add the Breathe HR video here with the video component, https://www.youtube.com/watch?v=jISXMPuO-Pk, then delete this line]
Step 1: connect Breathe HR to Tugger
First, copy your API key from Breathe:
- In Breathe, open Configure at the bottom of the menu and choose Settings.
- Under Integrations, click API Setup.
- Check the API shows as enabled.
- Copy your production API key.
Then add the connector in Tugger:
- Log in to Tugger, go to Connectors and choose Breathe.
- Give the connector a name and choose how much history to sync.
- Paste your Breathe API key.
- Check your credentials, then save.
Tugger runs a first sync and then refreshes on a schedule. You can see sync status and history in the Tugger dashboard.
Step 2: add Tugger's MCP server to Claude
- In Tugger, open AI Insights and copy your MCP server URL.
- In Claude, go to Settings, then Connectors.
- Choose Add custom connector.
- Name it "Tugger" and paste the MCP server URL.
- Click Add, then Connect, and sign in to Tugger when prompted.
In any new chat, open the connectors menu and switch Tugger on. Claude can now see the tables Tugger has synced from Breathe.
Step 3: ask your first question
Start with something simple to confirm the connection is working:
"Using my Breathe connector, show me how many days of holiday are booked by department for each week until the end of the year."
Claude will find the relevant tables, run the query and return the result. From there you can ask follow-ups, request a chart, or ask Claude to explain how it built the query. For people data, asking for results by department rather than by person keeps answers suitable to share.
What syncs from Breathe HR
| Data | What it covers |
|---|---|
| Employees and jobs | Employee records, job titles, employment type and contract dates |
| Departments, divisions and locations | How your organisation is structured |
| Absences and leave requests | Booked time off, pending requests, approval status and leave reasons |
| Sickness | Sickness absence, duration and fit note tracking |
| Holiday allowances and holiday years | Entitlement policies, carryover and usage per holiday year |
| Training | Courses, training types, cost, status and expiry dates |
| Expenses | Expense claims and individual expenses, with approval and payment status |
| Salaries, bonuses and benefits | Salary history, bonus awards, pensions and other benefits |
| Working patterns | Days and hours worked per pattern |
If you need an entity that isn't listed, ask us. Tugger's Breathe connector is extended based on customer requests.
Why go through a warehouse rather than connecting Claude to Breathe directly?
- Read-only by design. Some tools connect Claude straight to your Breathe account and let it carry out actions there. With Tugger, Claude can only read a copy of your data in the warehouse, so it can't change a record in Breathe.
- No load on Breathe. Questions run against the warehouse, not the Breathe API, so a year of absence history doesn't need pulling through the API each time you ask.
- One place for all your systems. With Breathe in the warehouse alongside Xero, HubSpot or Harvest, Claude can answer questions that span systems, such as headcount against revenue or utilisation.
Tugger is UK-based and ISO 27001:2022 certified. Your warehouse is hosted on Azure in the UK by default. The same warehouse works with Claude, ChatGPT and Gemini, and with Power BI and Tableau dashboards.
Related guides
- Breathe HR to Claude video walkthrough
- What is an MCP server?
- Why Claude needs a data warehouse
- How to connect Xero to Claude
- How to connect QuickBooks to Claude
- How to connect Sage 50 to Claude
- How to connect MYOB to Claude
- How to connect Pipedrive to Claude
- How to connect Cin7 to Claude
- How to connect Unleashed to Claude
- How to connect Shopify to Claude
- How to build a Claude dashboard
Get started
Connecting Breathe HR to Claude with Tugger takes no code and no faff. Try Tugger free for 10 days or book a demo and we'll walk through it with your own Breathe data.
