Hiring inside Claude means running ATS recruiting workflows — daily briefs, pipeline reviews, debrief prep, scorecard drafts, and permissioned stage moves — from inside Anthropic’s Claude assistant via a Model Context Protocol (MCP) plugin, without leaving the chat. Proof they can do the job is a role-specific task whose output a hiring manager can open, read, and defend in a debrief (a work sample or job simulation). Asking Claude “what needs me today?” or “draft this scorecard from my notes” answers “what is waiting in Greenhouse?” It does not by itself prove “this person can produce this job’s day-one deliverable on our JD.”
That shelf got a named product moment on 22 Oct 2026, when Greenhouse launched Greenhouse in Claude in open beta for Greenhouse Recruiting customers on Core, Plus, and Pro plans. The plugin runs on Greenhouse MCP, works in Claude, Claude Cowork, and Claude Code, and pairs a Greenhouse connector with skills for everyday hiring work: a daily brief, candidate status, debrief preparation (including who still owes a scorecard), pipeline review of stalled candidates and stage dwell time, scorecard drafts from interview notes, and write access to update candidates, move them forward, schedule interviews, and create jobs and job posts — all without leaving the conversation. Greenhouse remains the system of record; each person signs in with their own account so Claude can only see and do what they already can in Greenhouse; admins gate scopes. Nkem Nwankwo (VP Product Management, Greenhouse) and Mike Boufford (Head of People Products, Anthropic) both stress the same line in the newsroom release: people still make every hiring decision. Schmidt and Hunter (Psychological Bulletin, 1998) put work-sample tests near .54 predictive validity for job performance.
This guide is for TA leaders and hiring managers who hear “we hire from Claude now” or “the plugin already drafted the scorecard and moved them” and wonder whether ATS-in-LLM workflow replaces openable proof. It separates Claude-side recruiting ops from hire-time work evidence, and shows where Dotof AI Role-Simulation Assessments fit: after (or beside) the Claude brief, prove day-one work before you spend scarce panel time. Related (not rewrites): hiring orchestration, HCM recruiting optimizer agents, agentic hiring OS, and hiring companies.
What does Greenhouse in Claude actually deliver?
It delivers Greenhouse recruiting work where recruiters already chat with AI. Greenhouse’s 22 Oct 2026 newsroom post says teams can get answers from live hiring data and take approved actions in the same conversation, using the same structured hiring approach and permissions they already rely on in Greenhouse. The skills cover the ops day: what needs attention, where a candidate stands, who still owes feedback before a debrief, which applicants stalled and for how long, draft scorecards from notes, and confirmed write actions that update the ATS record. Claude waits for confirmation before changing anything in the Anthropic marketplace framing; Greenhouse MCP is the open-standard connector customers already use for approved AI tools.
Treat that as ATS workflow inside an LLM. Celebrate when recruiters stop tab-hopping to find overdue scorecards or stalled stages. Do not import “Claude prepared their debrief” or “the plugin drafted the scorecard and we moved them” into a new-hire decision as proof that Candidate A can produce this job’s day-one deliverable. The artifacts are a daily brief, a status summary, a debrief pack, a draft scorecard, and an ATS stage change. They are not a memo, a model, a plan, or a work product written against your live job description.
Why isn’t hiring inside Claude enough at hire time?
Because the hiring manager’s job in a debrief is to open evidence of work on this req — not a story about how quickly Claude assembled interview notes or advanced a stage. Greenhouse in Claude answers “what is waiting in the ATS, and what write action should I confirm?” A work sample answers “what would this person produce next week on our JD?” Managers defend shortlists to peers with the second answer. A candidate can clear every Claude-assisted status check, scorecard draft, and stage move and still leave the interview packet empty of role output.
Do not mix this with neighboring products. Hiring orchestration (Greenhouse’s 15 Oct 2026 Orchestrated Hiring Ecosystem) is the broader funnel of Dream Job flags, Real Talent identity, Voice AI transcripts, and the ATS. An agentic hiring OS is a co-pilot spanning sourcing, screening, assessments, and mock interviews. An HCM recruiting optimizer is engagement and rediscovery agents inside payroll HCM. Greenhouse in Claude is a named MCP plugin SKU: hire-ops skills and permitted writes inside Claude / Cowork / Code, with Greenhouse still system of record. It is not a task generated from your job description for every role you hire, and it is not a claim that Greenhouse or Anthropic buys or uses Dotof.
When should HR use a work sample instead of (or after) hiring inside Claude?
- Use Greenhouse in Claude for recruiter throughput when the bottleneck is status hunting, overdue scorecards, stalled pipelines, and context-switching out of the AI workspace. Keep a named human confirming every write — Greenhouse and Anthropic both say people still make every hiring decision.
- Add proof before panel time once the role has observable day-one deliverables, a wrong hire is costly, or managers refuse to interview without something they can open. Generate the task from the live JD — see from job description to role simulation.
- Prefer the sample over “Claude cleared them” as the hire gate for professional roles where “can they do the work?” is the question you cannot afford to get wrong. Use the plugin to decide what needs attention in Greenhouse. Use the sample to decide who advances on your JD.
Do not ask a Claude daily brief or a draft scorecard to answer “can they do Tuesday’s work on our req?” — that is a task, not an ATS status pack. Keep a written rule: no “moved in Claude” note should be the only hire vote. The practical gates are in the responsible AI hiring checklist. Finding and matching people is a separate product problem — see AI sourcing and matching — and it does not replace the artifact either.
Which should we buy for skills-based hiring in 2026 — a Greenhouse Claude plugin or a work sample?
Buy the sample for the hire gate — and treat Greenhouse in Claude as optional ATS-in-LLM workflow that still needs openable proof underneath. The plugin is a productivity purchase: daily brief → pipeline → debrief pack → draft scorecard → confirmed write actions inside Claude. A work sample is a selection purchase: openable evidence before panel hours. If your buyer question is “Greenhouse Claude plugin vs job simulation — which should we buy?”, start with the artifact managers will open in a debrief — then decide whether MCP hire-ops skills still earn their place for recruiter day-to-day. That is the answer to “hiring inside Claude vs job simulation”: put Greenhouse next to Claude if context-switching is the bottleneck; prove the work either way. See also AI interviewer vs job simulation, AI screening vs human judgment, and the anatomy of a useful AI assessment report for hiring managers.
How does Dotof fit this decision?
Dotof AI Role-Simulation Assessments start from the job description, generate a realistic task, run it in a supervised browser, and return structured evidence hiring managers can compare. Use Greenhouse in Claude when your pain is “recruiters live in Claude and keep losing time hunting status in the ATS.” Use the simulation when the decision you cannot afford to get wrong is whether this candidate can do the job — especially when the “evidence” so far was a Claude brief, a draft scorecard, or a confirmed stage move. LLM hire-ops and job-task proof can coexist. They should not be confused. Nothing in Greenhouse’s 22 Oct 2026 Claude plugin launch means Greenhouse or Anthropic buys, endorses, or uses Dotof.