About hiring-mender-mcp
hiring-mender-mcp is a local MCP server for recruiting workflows that use BambooHR or Greenhouse. It normalizes candidate and application write operations behind a shared set of tools.
Key features
- 01Create candidates and applications
- 02Advance candidates through stages
- 03Reject Greenhouse applications
- 04List valid statuses or job candidates
- 05Filter candidate lists by job
Use cases
- 01Add candidates to an ATS
- 02Move an application to another stage
- 03Reject a Greenhouse application
- 04Review candidates for a specific job
MCP support
The local MCP server exposes candidate and application management tools to MCP clients. It connects to BambooHR and Greenhouse APIs using supplied credentials.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
5Saved tool list · updated 1w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| platform | enum | Yes | bamboohr | greenhouse |
| candidate | object | Yes | Candidate fields. Field names differ by platform — see tool description. |
Details
From the project READMEHiring Mender MCP — BambooHR & Greenhouse ATS Write Tools for AI Agents
A free, local MCP server that creates candidates, advances/moves applications, and rejects candidates in BambooHR and Greenhouse — one consistent tool vocabulary across both platforms, so you don't hand-roll two different ATS integrations.
If you're building an AI recruiting workflow and keep rewriting the same candidate-creation/stage-advance logic for BambooHR's OAuth API and Greenhouse's Basic Auth Harvest API, this is the MCP server for that exact problem.
The problem this solves
BambooHR's Applicant Tracking API and Greenhouse's Harvest API both support real write operations — creating candidates, moving them through pipeline stages, rejecting applications — but they use completely different auth schemes (OAuth 2.0 bearer tokens vs. HTTP Basic Auth with a blank password) and completely different field names and endpoint shapes. There is no free, open-source tool that normalizes both behind one set of calls. This is that tool.
Tools
- create_candidate — create a new candidate/application. BambooHR:
firstName,lastName,jobId, etc. Greenhouse:first_name,last_name,job_id, etc. - advance_application — move a candidate to a new stage/status. BambooHR uses a numeric status ID (
list_statusesfirst). Greenhouse usesfrom_stage_id(required, guards against stale moves) andto_stage_id/to_job_id. - reject_application — reject a candidate (Greenhouse only; BambooHR rejections go through
advance_applicationwith a rejection status ID). - list_statuses — list valid BambooHR applicant status IDs, or Greenhouse candidates for a job.
- list_candidates — list candidates/applications, optionally filtered by job ID.
Setup
1. Install Node.js (v18+): https://nodejs.org
2. Clone and install:
git clone https://github.com/headlessherm-creator/hiring-mender-mcp.git
cd hiring-mender-mcp
npm install
3. Get API credentials:
- BambooHR: Requires an OAuth 2.0 access token with the
hiring:applications.writescope. See BambooHR's OAuth docs. - Greenhouse: Requires a Harvest API key, created by a user with "Can manage ALL organization's API Credentials" permission, via Configure → Dev Center → API Credential Management. See Greenhouse's Harvest API docs.
4. Set environment variables:
export BAMBOOHR_COMPANY_DOMAIN=yourcompany
export BAMBOOHR_ACCESS_TOKEN=your_oauth_token
export GREENHOUSE_HARVEST_API_TOKEN=your_harvest_token
You only need to set the variables for the platform(s) you're using.
5. Point your MCP client at it:
{
"mcpServers": {
"hiring-mender": {
"command": "node",
"args": ["/path/to/hiring-mender-mcp/src/index.js"],
"env": {
"BAMBOOHR_COMPANY_DOMAIN": "yourcompany",
"BAMBOOHR_ACCESS_TOKEN": "your_oauth_token",
"GREENHOUSE_HARVEST_API_TOKEN": "your_harvest_token"
}
}
}
}
Setup docs for common clients:
- Claude Desktop: https://modelcontextprotocol.io/quickstart/user
- Cursor: https://docs.cursor.com/context/model-context-protocol
- Claude Code: https://docs.claude.com/en/docs/claude-code/mcp
Security note
Greenhouse Harvest API keys are all-or-nothing within the scope you grant them — the key can access every endpoint type you enable when you create it, not just the ones this tool uses. Create a Harvest key with only the endpoints this tool needs (candidates, applications) rather than reusing a broad existing key. Same caution applies to the BambooHR OAuth scope — request only hiring:applications.write, not broader HR data access.
This server makes real network calls to BambooHR's and Greenhouse's APIs using the credentials you provide — unlike a fully local tool, your data does leave your machine, going directly to the ATS vendor you're already using. No third party, telemetry, or logging is involved.
Keywords
BambooHR MCP server, Greenhouse MCP server, ATS MCP server, applicant tracking system AI agent, recruiting automation MCP, HR MCP server, hiring pipeline automation, candidate management MCP, Claude MCP recruiting tools.
Related tools
Same local-first philosophy, different problems:
- polymath-megablaster-mcp — deterministic math, date, unit conversion, networking
- json-mender-mcp — fix malformed JSON, validate JSON Schema, test regex
- helpdesk-mender-mcp — Zendesk + Freshdesk + Help Scout write tools, one vocabulary
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.