Note: Greenhouse MCP is currently in open beta. Access depends on your subscription tier and admin configuration. Contact your account team to learn about access and pricing.
Greenhouse Model Context Protocol (MCP) lets approved AI tools connect directly to Greenhouse Recruiting so teams can work with live recruiting data inside the tools they already use. Once connected, AI assistants can surface insights, run analyses, answer questions, and complete supported actions using your Greenhouse data as context. There's no need to switch tabs or build custom reports.
MCP video overview
What is Greenhouse MCP?
Greenhouse MCP gives approved AI tools a secure, structured way to work with your hiring data. The MCP acts as a layer on top of the Greenhouse API, translating natural language requests into structured API calls, scoped to your existing permissions. This makes it possible to retrieve data, answer questions, and take supported actions without navigating the UI directly or building custom integrations.
Why use Greenhouse MCP?
Greenhouse MCP gives teams faster access to live recruiting data inside the tools they already use. Because it is built and maintained by Greenhouse, access is governed by your organization's existing permissions and admin-controlled scopes.
With Greenhouse MCP, teams can:
- Use live Greenhouse data inside approved AI tools without manual exports or custom integrations
- Keep access aligned with existing Greenhouse permissions and organizational governance
- Control which actions are available through admin-managed scopes in Dev Center > MCP Access
- Support auditability with Greenhouse-managed access controls and logging for supported MCP activity
MCP use cases
Greenhouse MCP is useful for recruiting operations, pipeline visibility, and hiring analysis. Representative examples include:
- Building a live pipeline dashboard for hiring managers across active jobs
- Seeing how long candidates have been in each stage and flagging stalled candidates
- Identifying recruiter bottlenecks across open jobs
- Checking scorecard completion and consistency across hiring managers
- Comparing interview kits across jobs to find missing or inconsistent questions
- Auditing jobs that are missing a job description or posting
- Analyzing offer approval bottlenecks by approver
For a full list of examples with sample prompts, read about Greenhouse MCP use cases by role.
Before you begin
Before connecting Greenhouse MCP, confirm the following:
- Your organization is on GHR Core, Plus, or Pro
- You sign in using the email address tied to your Greenhouse Recruiting account
Supported AI tools
Greenhouse MCP currently supports:
- Claude
- ChatGPT
- Gemini CLI
- Amazon Q
- Grok AI
Note: Gemini is only supported through Gemini CLI. The Gemini web app does not currently support connecting to remote MCP servers like Greenhouse MCP.
Any AI tool that supports OAuth 2.0 with PKCE authentication and Dynamic Client Registration can also work with Greenhouse MCP. To connect a different tool, share the redirect URL with the Greenhouse team so it can be added to the allowlist. This step is required for the authentication flow to complete successfully.
Connect Greenhouse MCP
To connect Greenhouse MCP, a Site Admin starts the flow in a supported AI tool and enables the required scopes during setup. Users then authorize access using the email address tied to their Greenhouse Recruiting account. Once connected, the AI tool can use the capabilities approved for the organization.
For full, tool-specific setup instructions, read about Greenhouse MCP use cases by role.
Tip: If testing in a Greenhouse sandbox environment, log out of your account first before starting the connection flow. This helps avoid connecting to the wrong environment or account.
Manage access and scopes
Site Admins manage what their team can access in Dev Center > MCP Access. This is where admins review and enable the scopes users can approve when connecting an AI tool. Scopes set an upper limit—users cannot grant themselves access beyond what the admin has enabled.
Access is governed by three factors:
- Scopes approved by the organization in Dev Center > MCP Access
- The user's existing Greenhouse permissions
- Greenhouse MCP supported capabilities
If scopes are updated, users will need to reconnect their account for the changes to take effect.
Get better results
Greenhouse MCP uses live Greenhouse data. The cleaner and more structured that data is, the better the outputs. If you're not sure where to start, begin with interview stages and scorecards—these two areas tend to have the biggest impact on output quality.
Keep interview stages consistent
Use clear and consistent interview stage names across jobs so MCP can analyze pipeline movement more accurately.
Complete scorecards and interview kits
Structured scorecards, consistent recommendations, and detailed notes give MCP better context for candidate summaries, debriefs, and pattern analysis.
Keep candidate activity up to date
Move candidates through stages promptly, reject candidates who are no longer active, and capture clear rejection reasons. This helps MCP identify delays, bottlenecks, and drop-off patterns.
Keep job setup complete and accurate
Active jobs should include accurate hiring teams, office and department details, and up-to-date job information. This improves job-level and team-level analysis.
Use structured custom fields where possible
Fields like single select, multi-select, yes/no, number, date, and currency are easier for MCP to analyze than free-text fields.
Review permissions carefully
Greenhouse MCP respects existing Greenhouse permissions. Thoughtful permission setup helps users get the access they need while keeping recruiting data secure.
Troubleshooting
I don't have access to Greenhouse MCP
If you can't access Greenhouse MCP, confirm the following:
- Your organization is on GHR Core, Plus, or Pro
- A Site Admin has configured the required scopes in Dev Center > MCP Access
- The AI tool supports OAuth 2.0 with PKCE authentication and Dynamic Client Registration
- The redirect URL for the AI tool has been allowlisted by Greenhouse, if required
I signed in with the wrong account
Disconnect the Greenhouse connector in your AI tool, sign out of your Greenhouse account, and start the connection again using the email address tied to your Greenhouse Recruiting account.
If switching between sandbox and production environments, log out of the current environment before reconnecting.
My AI tool is missing capabilities I expected
Greenhouse MCP only exposes capabilities that are supported today and approved for your organization. If a capability is missing, check the following:
- Whether the required scopes are enabled in Dev Center > MCP Access
- Whether your Greenhouse user permissions allow the action
- Whether a reconnect is needed if scopes were recently updated
Note: Greenhouse MCP is in open beta, and supported capabilities will continue to expand over time.
Frequently asked questions
Who can use Greenhouse MCP?
Greenhouse MCP is available to Site Admins on GHR Core, Plus, and Pro subscription tiers.
What if my organization is on a legacy tier?
Greenhouse MCP is only available for GHR Core, Plus, and Pro tiers. If your organization is on a legacy tier, contact your account team to learn about access options.
Is Greenhouse MCP generally available?
Not yet. Greenhouse MCP is currently in open beta.
Which AI tools are supported?
See Supported AI tools above for the current list and information on connecting other tools.
Does access follow my Greenhouse permissions?
Yes. Greenhouse MCP is designed to work with your organization's existing Greenhouse permissions and admin-approved scopes. Users cannot access data or take actions beyond what their Greenhouse permissions and organization-approved scopes allow.
Can Greenhouse MCP take actions, or is it read-only?
Greenhouse MCP supports both read and write actions—it is not limited to read-only access. Whether a specific action is available depends on what Greenhouse MCP supports today, which scopes the organization has enabled in Dev Center > MCP Access, and the user's existing Greenhouse permissions.
What if I approve the wrong scopes, or my organization updates scopes later?
Disconnect the Greenhouse connector in your AI tool, reconnect, and complete the authorization flow with the correct scopes. Newly added scopes are not applied automatically—users must reconnect to pick them up. If the organization removes scopes, all MCP users are signed out. See Troubleshooting above for additional guidance.