Open Greenhouse MCP Server
Greenhouse recruitment data access with configurable roles and bulk workflow tools.
Data last scanned today · Reviewed today
Overview
Open Greenhouse MCP Server streamlines hiring team operations by providing specialized profiles for recruiters. Unlike standard API mirrors, this server groups Greenhouse actions into workflow-oriented profiles including read-only reporting, recruiter-specific daily tasks, and full administrative access.
Our verdict
This server carries a high-risk rating of 50/100, primarily because it is a community-maintained project rather than an official Greenhouse release. While it includes valuable safety features like a read-only configuration and explicit ID requirements for destructive actions, users should exercise caution given the lack of independent editorial review. It is recommended for organizations that have thoroughly reviewed the source code and can strictly control the API permissions associated with their Greenhouse credentials.
- Includes a dedicated read-only configuration profile
- Zero direct dependencies reduce supply-chain surface area
- Supports audit attribution for write operations
- Lacks support for OAuth authentication
- Project is community-maintained with no official vendor support
- Not yet undergone formal editorial review
Tools
| Tool | Description | Risk |
|---|---|---|
| screen_candidate | Complete screening package — profile, resume text, location, screening answers, job description, history | low |
| fetch_new_applications | Applications since a date, grouped by job — the daily recruiter digest | low |
| scan_pipeline_resumes | Search resume text across pipelines with boolean keywords (required/preferred/exclude) | low |
| search_pipeline_candidates | Search pipelines by structured fields — title, company, education, experience, tags | low |
| scan_all_candidates | Database-wide candidate search by structured fields with date bounds | low |
| batch_read_resumes | Batch-fetch and extract resume text for multiple candidates | low |
| pipeline_summary | Full pipeline view — candidates grouped by stage with names and days-in-stage | low |
| candidates_needing_action | Find stale applications and interviews missing scorecards | low |
| stale_applications | Applications with no activity for N days, sorted by stalest | low |
| pipeline_metrics | Conversion rates, hire/rejection rates, time-in-stage per stage | low |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf | ||||
| ChatGPT |
Frequently asked questions
›What are the different tool profiles available?
The server offers three profiles: read-only for reporting and hiring managers, recruiter for daily tasks, and full for administrative operations.
›How do I set up my credentials?
You need to generate a Harvest API key in your Greenhouse account under Configure, Dev Center, and API Credential Management, then add it to your MCP client config.
›Does this server support bulk operations?
Yes, it includes tools for bulk actions like rejecting or tagging candidates, which are rate-limited to ensure they stay within API limits.
›How can I verify my configuration during startup?
The server logs its active profile, the total number of available tools, and enabled API categories whenever it starts.
›What is the recommended approach for initial deployment?
It is recommended to start using the read-only profile to validate connectivity and tool behavior before switching to profiles that allow write access.
›How does the server handle audit tracking for write operations?
Write operations support audit attribution through the GREENHOUSE_ON_BEHALF_OF environment variable.
Badge
Maintain this server? Add the live badge to your README.