Greenhouse MCP Server
Greenhouse is an applicant tracking system (ATS) and recruiting platform. This connector targets the Harvest v3 API for managing candidates, applications, jobs, interviews, scorecards, and offers.
What is the Greenhouse MCP Server?
The Greenhouse MCP server gives AI agents structured, permission-aware access to Greenhouse through the Model Context Protocol. With 57 pre-built actions, agents can read, create, and update Greenhouse data on behalf of authorized users.
Willow ships the Greenhouse MCP server as part of an enterprise control plane. Every call runs behind SSO (Okta, Azure AD), enforces RBAC and least-privilege at runtime, writes to a full audit trail, and integrates with Splunk and Loki for SIEM visibility. Connect from Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code, n8n, or any custom agent. Install once, distribute org-wide, and see exactly how Greenhouse is being used by every AI agent in your stack.
Tools
List Candidates
List candidates and prospects in the Greenhouse account. Use cursor-based pagination via `cursor` (do not combine with other filters). Filter by `ids`, by `email`, by a `tag` name, or by date with `created_at`/`updated_at` (operator-prefixed values, e.g. `gte|2024-01-01T00:00:00Z`).
Create Candidate
Create a new candidate in Greenhouse. Provide at least `first_name` and `last_name`; supply an `application` object to also create the candidate's first application against a job. To create a Prospect instead, set `application.prospect = true`.
Update Candidate
Update an existing candidate's profile fields. Only the supplied fields are changed; arrays like `email_addresses` replace the existing collection. Recruiter and Coordinator are managed on the application object, not here.
Delete Candidate
Permanently delete a candidate (and all their applications) from Greenhouse. This is irreversible — prefer Anonymize for GDPR-style erasure that preserves recruiting history.
Merge Candidates
Merge two duplicate candidate records. The `primary_candidate_id` survives; the duplicate is merged into it and then removed. Applications, activity, and tags are consolidated on the primary.
Anonymize Candidate
Anonymize a candidate's personal data for GDPR/privacy compliance. Specify which `fields` to anonymize (e.g. `full_name`, `phone_numbers`, `emails`, `social_media_links`, `addresses`, `notes`, `rejection_notes`, `email_addresses`, `activity_items`, `attachments`). Unlike delete, anonymize preserves the recruiting record in aggregate reporting.
List Candidate Tags
List all candidate tags defined in the Greenhouse account. Tags are org-wide labels that can be applied to any candidate.
Create Candidate Tag
Create a new org-wide candidate tag. Once created, it can be applied to any candidate.
Apply Candidate Tag
Apply an existing tag to a candidate. The tag must already exist (use Create Candidate Tag first if needed).
Remove Candidate Tag
Remove a tag from a candidate. The tag itself remains in the org and can be re-applied later.
List Applications
List applications across the account. Filter by parent (`candidate_ids`, `job_ids`, `job_post_ids`, `source_ids`, `stage_ids`), by `status` (`active`/`hired`/`rejected`/`converted`), by `stage_name`, or by date with `created_at`/`updated_at`/`last_activity_at` (operator-prefixed). Use `cursor` to walk pages.
Move Application
Move an application forward in the pipeline. Provide `from_stage_id` (the stage you're moving out of). Set `to_stage_id` to choose a destination stage in the same job; add `to_job_id` to transfer to another job. Leave both blank to advance one stage.
Hire Application
Mark an application as hired. Optionally set the `start_date`, the `opening_id` to fill, and a `close_reason_id` for any related opening that's being closed.
Reject Application
Reject an application with a specific rejection reason. Optionally include a `notes` blurb, a scheduled `rejection_email` payload, and any `custom_fields` collected during rejection.
Unreject Application
Reverse a rejection on an application, returning it to the active pipeline at its prior stage.
Create Application
Create a new application (or prospect) for an existing candidate. Set `prospect: true` in the body to create a Prospect instead of a Candidate application. Provide `job_id` to apply to a specific job.
Update Application
Update an existing application's source, referrer, prospect fields, or custom fields. Only the supplied fields are changed.
Delete Application
Permanently delete an application. This is irreversible — the candidate record may remain but the application and its associated data are removed.
Convert Prospect
Convert a prospect application into a candidate application on a specific job. The prospect record is converted in-place.
List Application Stages
List the job interview stages (pipeline stages) across the account. Each stage belongs to a job and represents a step in its hiring pipeline, with a human-readable `name` (e.g. `Phone Screen`, `Onsite`). Filter by `job_ids`, by specific `ids`, or by `active` to restrict to a job's current interview plan.
List Job Interviews
List job interviews (the named interview slots/templates configured on a job's interview plan, e.g. `Demo & System Design Interview`). Each interview belongs to a job interview stage and carries the human-readable `name` referenced by a scheduled interview's `job_interview_id`. Filter by `job_ids`, `job_interview_stage_ids`, `active`, or `scheduling_type`.
List Jobs
List jobs in the account. Filter by `status` (`open`/`draft`/`closed`), `requisition_id`, `department_id`, `office_id`, `confidential` flag, by `opened_at`/`closed_at`/`created_at`/`updated_at` operator filters, or by specific `ids`. Use `cursor` for pagination.
Create Job
Create a new job by cloning a template job. `template_job_id` and `number_of_openings` are required; the rest of the fields override values copied from the template.
Update Job
Edit a job's metadata: name, notes, anywhere flag, requisition ID, hiring narrative fields (`team_and_responsibilities`, `how_to_sell_this_job`), department/office assignment, and custom fields. Only the provided fields are written.
List Job Posts
List job posts (career-site listings). Filter to `live: true` to only see what's published on a live job board, by `job_ids` to fetch posts for specific jobs, by `internal: true` for internal-only posts, or by `featured`/`active`. Use `cursor` for pagination.
Create Job Post
Create a new job post (career-site listing) for a job. The post is created in draft; set `active: true` to publish it immediately.
Update Job Post
Update an existing job post. Only the supplied fields are changed. Set `active: false` to unpublish or `active: true` to re-publish.
List Openings
List openings (head-count slots on a job). Filter by `job_ids`, `status` (`open`/`closed`), by specific `ids`, or by date filters. Use `cursor` for pagination.
Create Opening
Create a new opening (head-count slot) on a job. Greenhouse caps total open openings at 100 per job.
Update Opening
Update an existing opening. Change its `status` or `close_reason_id`, or update custom fields.
List Interviews
List scheduled interviews. Filter by `application_ids`, `job_ids`, `organizer_ids`, `job_interview_ids`, `external_event_id`, by lifecycle `status`, or by `starts_at`/`ends_at`/`all_day_*` operator filters. Use `cursor` for pagination.
List Scorecards
List interview scorecards. Filter by `application_ids`, `interviewer_ids`, `submitter_ids`, or `interview_kit_ids`; by `status` (`draft`/`complete`); or by `interviewed_at`/`submitted_at`/`created_at` operator filters.
List Interviewers
List per-interview interviewer assignments. Filter by `interview_ids`, `user_ids`, `scorecard_ids`, or `response_status` (`needs_action`/`accepted`/`declined`/`tentative`). Useful for working out who should still leave feedback.
Create Interview
Schedule a new interview for an application. Provide `application_id`, `job_interview_id` (the interview step from the pipeline), and a list of `interviewers` with their user IDs.
Update Interview
Update an existing scheduled interview. Change the time, location, interviewers, or external event link.
Delete Interview
Delete a scheduled interview. This removes the interview record and any associated scorecard assignments.
List Interview Kits
List interview kits (the questionnaires/rubrics attached to each interview step). Filter by `job_ids` to see kits for specific jobs, or fetch by `ids`.
List Offers
List offers. Filter by `application_ids`, `candidate_ids`, `job_ids`, or `opening_ids`; by `status` (`Created`/`Accepted`/`Rejected`/`Deprecated`); or by `sent_on`/`starts_on`/`resolved_at` operator filters. Set `current_only: true` to only get each application's most recent offer.
Create Offer
Create a new offer for an application. Provide `application_id` and the offer details. The offer is created in `Created` status; sending it is a separate step in Greenhouse UI.
Update Offer
Update an existing offer. Change its status, dates, or custom fields.
List Notes
List notes and activity entries. Filter by `candidate_ids`, `application_ids`, `note_type` (`NOTE`/`ACTIVITY`/`EMAIL`), or by `user_ids` (who wrote the note). Use `cursor` for pagination.
List Attachments
List attachments (resumes, cover letters, admin uploads). Filter by `candidate_ids`, `application_ids`, or `type` (`resume`/`cover_letter`/`admin_only`/`take_home_test`/`offer_packet`/`other`).
Download Attachment
Download an attachment's file, such as a resume or cover letter. Pass the `url` that List Attachments returned, unchanged. It is a signed link that expires after 7 days; list the attachments again for a fresh one. PDF and Word files come back as text, alongside the file itself.
Create Attachment
Upload an attachment to a candidate. Provide the file as a base64-encoded `content` string, a `filename`, and a `type` (`resume`, `cover_letter`, `admin_only`, etc.).
Delete Attachment
Delete an attachment by ID. This is irreversible.
List Users
List Greenhouse users (recruiters, hiring managers, etc.). Filter by `primary_email`, `office_ids`, `department_ids`, `linked_candidate_ids`, `interviewer_tag_ids`, `agency_ids`, `employee_ids` (string IDs), `external_office_id`, or `external_department_id`. Set `deactivated: true` to include only deactivated users, or `show_service_accounts: true` to include integration service users.
List Offices
List offices in the account. Optionally filter to a `parent_id` (for hierarchical lookups), an `external_id`, or specific `ids`. Pagination via `cursor`.
List Departments
List departments in the account. Optionally filter to a `parent_id`, an `external_id`, or specific `ids`. Pagination via `cursor`.
List Sources
List candidate sources (e.g. LinkedIn, Internal Referral, Sourcer). Useful to map a `source_id` from an application back to a human-readable label.
List User Roles
List user roles configured on the Greenhouse account (Site Admin, Job Admin, Interviewer, etc.). Use to translate a user's `role_id` into a human-readable role.
List Custom Fields
List custom field definitions configured on the account. Filter by `field_type` (`candidate`/`application`/`job`/`offer`/`opening`/`rejection_question`) to narrow to a resource type.
List Custom Field Options
List the selectable options for single-select or multi-select custom fields. Provide `custom_field_id` to get options for a specific field.
List Rejection Reasons
List rejection reasons configured in the account. These are used when rejecting an application via Reject Application.
List Close Reasons
List close reasons (used when closing a job or an opening). Useful to populate drop-downs in hire/close workflows.
List Email Templates
List email templates available in the account. Templates are used for rejection emails, offer letters, and other candidate communications.
Get Bulk Request
Check the status of an asynchronous bulk request (e.g. a bulk import or update). Returns `status` (`queued`/`processing`/`complete`/`error`), progress counts, and any error details.
Create Note
Add a note, activity entry, or email log to a candidate. `note_type` chooses the kind (`NOTE`, `ACTIVITY`, or `EMAIL`); `visibility` is one of `public`, `private`, `admin_only`. Provide `candidate_id` (and optionally `application_id` to attach to a specific application). For `EMAIL`, set `email_from`/`email_to`/`email_cc`.
Customize Tools
Edit descriptions, modify arguments, select tools, or add new ones
Set Up Your Greenhouse MCP Server in Minutes
Add the following configuration to your MCP client. Authentication is handled via OAuth. Compatible with Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code, n8n, and any MCP-compatible agent.
Claude Desktop
{
"mcpServers": {
"willow-greenhouse": {
"type": "http",
"url": "https://<org>.mcp-s.com/mcp/mcp/greenhouse"
}
}
}Cursor
{
"mcpServers": {
"willow-greenhouse": {
"type": "http",
"url": "https://<org>.mcp-s.com/mcp/mcp/greenhouse"
}
}
}Claude Code
claude mcp add willow-greenhouse --transport http https://<org>.mcp-s.com/mcp/mcp/greenhousen8n
{
"url": "https://<org>.mcp-s.com/mcp/mcp/greenhouse",
"method": "POST"
}Or click "Install with Willow" above to set up automatically with SSO and RBAC preconfigured.
Enterprise Governance for Greenhouse
Willow adds the layer Greenhouse and every other SaaS doesn't ship out of the box: every call runs behind SSO (Okta, Azure AD), enforces RBAC and least-privilege at runtime, writes to full audit logs, and detects shadow AI usage across your stack. One MCP gateway. Any agent. Every tool.
Greenhouse MCP Server FAQ
What is the Greenhouse MCP server?
The Greenhouse MCP server is a Model Context Protocol implementation that lets AI agents like Claude, Cursor, and ChatGPT read and write Greenhouse data through a standardized interface. Willow hosts and governs this server so enterprises can roll it out without a security review backlog.
How is Willow's Greenhouse MCP server different from the official one?
The official Greenhouse MCP server is scoped to a single user's account and does not include enterprise governance. Willow's version adds SSO, RBAC, audit logging, shadow AI detection, and centralized control over which actions agents can take across the entire org.
Which AI clients work with the Greenhouse MCP server?
Claude Desktop, Claude Code, Cursor, ChatGPT, VS Code with MCP support, n8n, and any custom agent built with OpenAI Agents SDK, LangChain, Vercel AI SDK, or Anthropic SDK.
Is the Greenhouse MCP server secure? How does Willow handle authentication?
Every call runs behind your existing SSO (Okta, Azure AD). Per-user OAuth scopes the agent to exactly what that user can do in Greenhouse, nothing more. No credentials reach the LLM. Every action writes to an audit trail.
Can I limit which Greenhouse actions agents can take?
Yes. Willow lets you scope agents to specific actions, specific projects, or specific environments. Toggle actions on or off in the dashboard, or enforce policy via infrastructure-as-code through GitHub.
How do I detect shadow Greenhouse MCP servers in my org?
Willow's browser extension and discovery service surface unmanaged MCP servers, skills, and AI agents across the org. If a developer installed an unapproved Greenhouse MCP locally, you'll see it.
What does the Greenhouse MCP server cost?
Pricing depends on org size and deployment model (SaaS, dedicated cloud, self-host). See withwillow.ai/pricing or contact sales for a quote.
How do I install the Greenhouse MCP server with Willow?
Install via the Willow Connect Panel in one click, or paste the JSON snippet above into your Claude Desktop, Cursor, or Claude Code config. SSO and RBAC inherit from your existing Willow setup.
Compare Willow MCP Gateway
See how Willow stacks up against other MCP platforms on governance, security, and enterprise readiness.
Your agents are already in the wild.
Give them a Basecamp. Go from AI chaos to AI work, in minutes.