Integration · Payroll & HR
Connect Claude & ChatGPT to Greenhouse
Ask your AI which applications are waiting in a stage, what interviewers wrote in their scorecards, or to move a candidate forward in Greenhouse. Through the Harvest API it can read jobs, applications, candidates, interviews, offers, scorecards and attachments. It can also create and update candidates and move, reject, unreject or hire applications.
Free plan, no credit card. Takes about a minute.
Try asking
“Which active applications for the Product Designer job are in the Onsite stage?”
The AI finds the job and its stages and lists active applications in that stage.
“Summarize the scorecards for Julie's application.”
It reads the interview scorecards with ratings and feedback and summarizes them.
“Move Julie to the Offer stage.”
It moves the application from its current stage to the stage you name.
What your AI can do
Jobs and pipeline
List jobs, job posts, stages, departments and offices, and see applications by job, stage and status.
Candidates and feedback
Read candidates, attachments, interviews, scorecards and offers, and create or update candidates and tags.
Decisions
Move applications between stages or jobs, reject or unreject them, and mark an application as hired.
“Which offers are out right now?”
It lists current offers with their status.
“Add Peter Holm from Acme ApS to the Data Analyst job and tag him Referral.”
It creates the candidate with company and contact details, adds an application to the job and adds the tag.
“What interviews are scheduled this week?”
It lists scheduled interviews in the date range.
22 tools for Greenhouse Harvest
These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.
create_candidateChanges dataCreate a candidate. Required: first_name and last_name. Optional: email_addresses, phone_numbers, company, title, tags, and application with job_id to add them to a job.hire_applicationChanges dataMark an application as hired. Required: application_id. Optional: start_date, opening_id, close_reason_id. This closes the opening.list_applicationsList applications. Optional: candidate_ids, job_ids, stage_ids, status (active, rejected, hired...), and date filters.list_attachmentsList candidate attachments such as resumes and cover letters. Optional: candidate_ids, application_ids, type.list_candidate_tagsList the candidate tags in the account.list_candidatesList candidates. Optional: email, tag, ids, and created_at/updated_at/last_activity_at filters.list_departmentsList departments.list_interviewsList scheduled interviews. Optional: application_ids, job_ids, status, starts_at/ends_at filters.list_job_postsList job posts (public listings). Optional: job_ids, live, active, internal.list_job_stagesList a job's interview stages. Optional: job_ids, active.list_jobsList jobs. Optional: status (open, draft, closed), department_id, office_id, ids, and date filters like opened_at.list_offersList offers. Optional: application_ids, candidate_ids, job_ids, status, current_only.list_officesList offices.list_rejection_reasonsList the rejection reasons and their IDs.list_scorecardsList interview scorecards with ratings and feedback. Optional: application_ids, interviewer_ids, status, submitted_at.list_sourcesList candidate sources (job boards, referrals, agencies...).list_usersList Greenhouse users. Optional: primary_email, department_ids, office_ids, deactivated.move_applicationChanges dataMove an application to another stage. Required: application_id and from_stage_id; set to_stage_id, or to_job_id to move it to another job.reject_applicationVisible to othersReject an application. Required: application_id and rejection_reason_id. Optional: notes and rejection_email (email_template_id, send_email_at) to email the candidate.tag_candidateChanges dataAdd a tag to a candidate. Required: candidate_id and candidate_tag_id.unreject_applicationChanges dataUndo a rejection and put the application back in its pipeline. Required: application_id.update_candidateChanges dataUpdate a candidate's name, company, title, contact details, tags or custom fields. Required: candidate_id.
Set up in three steps
- 1
Pick the app
Create a free PipMCP account and choose Greenhouse Harvest from the app list.
- 2
Paste your key
In Greenhouse go to Configure → Dev Center → API Credential Management and create a Harvest v3 (OAuth) credential, ideally tied to an Integration System User. Choose its permissions, then copy the Client ID and Client secret into PipMCP. Greenhouse shows the secret only once.
- 3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
Click your name, then Settings › Connectors › Add custom connector. Paste your link as the Remote MCP server URL. Switch it on from the + menu in a chat.
Questions
What can the AI do in Greenhouse?
The AI can read jobs, job posts, stages, applications, candidates, attachments, interviews, scorecards, offers, sources, tags, rejection reasons, departments, offices and users. It can create and update candidates, tag them, move applications, reject or unreject them, and mark an application as hired.
Does the AI see my Greenhouse credentials?
No. Your Greenhouse client secret is encrypted at rest and never shown to the AI. After you save it, it is not shown again, not even to you. The AI only sees the results of the tools it calls.
Can I control what the AI is allowed to do?
Yes. You choose which tools are switched on, and the permissions on the Harvest credential limit what can be reached. Candidate data is personal data, so you can start read-only and keep reject_application off if a person should make every rejection. Hiring closes the opening, and it can require your approval before it runs. Every tool call is logged.
Does it work with ChatGPT?
Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.
What does it cost?
Greenhouse connects with OAuth, and OAuth connections need the PipMCP Pro plan or above. PipMCP also has a free plan with no credit card, but it does not include this connection. Paid plans bill per completed task.
Can the AI send rejection emails to candidates?
Only if you let it. reject_application can include a rejection email template and a send time, and then Greenhouse emails the candidate. Leave the tool off, or require approval for it, if you want to check every rejection first. unreject_application puts an application back in its pipeline.
Let your AI work in Greenhouse Harvest today.
Start free. Your key stays encrypted, and you decide what the AI may do.
Connect Greenhouse Harvest freePipMCP is not affiliated with Greenhouse Harvest. Product names are trademarks of their owners.