Connect AI assistants like ChatGPT and Claude to your Teamtailor account using the Model Context Protocol (MCP). Once connected, you can use your AI assistant to interact with Teamtailor using natural language. Ask questions, complete recruitment tasks, and streamline everyday workflows based on your existing Teamtailor permissions.
A Company Admin user needs to activate this feature in the Add-on feature center to get started
What is MCP?
The Model Context Protocol (MCP) is an open standard that enables AI assistants to securely communicate with external systems like Teamtailor.
Once connected, compatible AI assistants can retrieve information and perform actions in Teamtailor on behalf of your user, depending on the permissions granted by your organization. For example, it can answer questions about candidates and jobs, summarize recruitment activity, help prepare interviews, or automate repetitive recruitment tasks.
Authentication is handled through OAuth, meaning you sign in with your own Teamtailor account without sharing passwords or API keys with the AI assistant.
Looking for AI features built into Teamtailor?
Check out Co-pilot, our own AI assistant integrated directly into Teamtailor.
Set up MCP
Once the MCP feature has been activated in the Add-on feature center, the configuration can be found under Settings → Integrations → MCP. Here, the Company Admins can control:
Which user roles are allowed to connect AI assistants
Whether each role has read, create & modify, or delete permissions
Server URL
Use the following URL to connect your MCP client to Teamtailor:
https://mcp.teamtailor.com/mcp
Permissions
Every AI assistant connects through an individual Teamtailor user account and can only access data and perform actions that the connected user is already permitted to access in Teamtailor. AI assistants never receive their own permissions.
Here you can choose which Teamtailor user roles are allowed to use MCP and what level of access AI assistants should have for each role.
The permissions configured here act as an additional layer of control. An AI assistant can only perform actions that are permitted by both the connected user's Teamtailor permissions and the access level configured for their role. By default, all roles have Read access enabled, but you can modify these permissions at any time.
See the full list of actions available here.
Security
AI assistants authenticate using your Teamtailor account and never receive their own credentials or permissions. Access is limited by:
Your Teamtailor role
The MCP permissions configured by your Company Admin
OAuth authentication
All supported actions are performed on behalf of the signed-in user and are recorded in the Audit log. This allows you to see what actions were taken, when they occurred, and which Teamtailor user account the assistant was connected through.
Connect an AI assistant
Teamtailor provides a standard remote MCP server that can be connected to any compatible MCP client. The setup examples below use ChatGPT and Claude:
ChatGPT
Go to Settings → Apps → Advanced settings in your ChatGPT account and toggle on Developer mode.
Go back to Apps and click Create app.
Fill in the following information:
Name: Teamtailor
Connection: Select Server URL and enter
https://mcp.teamtailor.com/mcpAuthentication: Select OAuth
Select "I understand and want to continue" and click Create.
Sign in with Teamtailor to connect your workspace.
Claude
Navigate to Organization settings → Connectors in your Claude account.
Click the Add button.
Hover over Custom, then select Web.
Add the Teamtailor MCP server URL:
https://mcp.teamtailor.com/mcpOptionally, click Advanced settings to specify an OAuth Client ID and OAuth Client Secret for your server.
Finish the configuration by clicking Add.
Sign in with Teamtailor to connect your workspace.
Use AI assistant with Teamtailor
You can ask your AI assistant in natural language to:
Answer questions about your jobs, candidates, and recruitment processes.
Summarize candidate profiles, interview feedback, comments, and meetings.
Search across jobs, applications, meetings, messages, and other recruitment data.
Help prepare interviews or create summaries for hiring managers.
Automate repetitive recruitment tasks using your existing Teamtailor data.
Perform supported actions, such as updating recruitment data, if your Company Admin has granted your role
Create & modifyaccess
Examples:
What draft jobs do I have?
Can you give me an overview of how job ID 123 is configured?
Move John Doe to the Offer stage.
Remind me to follow up with the candidate tomorrow.
Rewrite the pitch for job ID 123 to make it more engaging.
Supported actions
Depending on your Teamtailor role and the MCP permissions configured by your Company Admin, you'll have access to one or more of the access levels below.
Read
Action | Description |
| List jobs, filter by status or your own jobs |
| View full job details, including content, application form, and recruitment process |
| View the recruitment pipeline, and candidate counts for one or more jobs |
| Search candidates by name, email, tags, status, job, stage, and more |
| View a candidate's profile, applications, files, competencies, and recent activity |
| List job applications and filter by job, stage, or rating |
| View detailed application information, including scorecards, comments, messages, and activity |
| View a candidate's message history |
| View internal comments on a candidate |
| View scorecards and evaluations for a candidate |
| View a candidate's activity |
| List upcoming and past candidate meetings |
| View meeting details, participants, and Co-pilot summaries |
| View the transcript from a meeting |
| List questions |
| View a question, including answer options and translations |
| List reusable interview kit templates |
| List available scorecard skills and traits |
| List to-dos across jobs, candidates, and applications |
| List existing candidate and job tags |
| List custom fields and their available options |
| List departments |
| List divisions |
| List roles |
| List locations |
| List employees and filter by role, department, division, or location |
Create & modify
Action | Description |
| Update job details such as title, description, and application form settings |
| Move a candidate to another stage in the recruitment process |
| Create a new question |
| Update an existing question |
| Restore an archived question |
| Add a question to a job's application form |
| Attach an interview kit to a job or stage |
| Add a skill or trait to a job's scorecard |
| Create a new to-do with an optional due date and assignee |
| Update a to-do's text, due date, or assignee |
| Mark a to-do as complete (or reopen it) |
| Add a tag to a candidate or job |
| Set, update, or clear a custom field value on a candidate or job |
Delete
Action | Description |
| Archive a question |
| Remove a question from a job's application form |
| Remove an interview kit from a job |
| Remove a skill or trait from a job's scorecard |
| Remove a tag from a candidate or job |
| Delete a to-do |
Legal
Teamtailor provides the MCP server but does not provide or control the external AI assistant or other MCP client that the user chooses to connect. The connected service is a separate third-party service and is subject to its provider’s own terms, privacy notice and data-handling practices including whether or not data may be used for training purposes.
Once data reaches the MCP client, its handling of data is governed by that provider's own terms and conditions, and privacy policy, and by the settings you have with them. Review those before connecting.
The customer is solely responsible for determining which external AI clients are permitted to connect, for ensuring that its selection, configuration and use of the AI service complies with the EU AI Act and/or any other laws or regulatory requirements applicable to the customer, and for informing its users of the applicable internal policies.
Read more in our Teamtailor MCP Server Privacy Policy here.
Prefer reading in your own language? Check out our other support centers available at the top of the page 🌍