Introducing Xobin MCP Server
Xobin now supports MCP, letting Claude, ChatGPT and other AI assistants securely search hiring evidence and act on recruiting workflows.
Aman Kumar Tiwari
Xobin MCP is live. Connect a compatible AI assistant, ask for the hiring evidence you need, and act on it without leaving the conversation.
Today we are releasing the Xobin MCP Server. It uses the Model Context Protocol, the open standard for connecting AI applications to external systems, so an authorized assistant can reach supported, live recruiting context in your Xobin account instead of relying on pasted profiles and stale exports.
You ask in natural language. Xobin supplies the hiring context, evidence and supported workflow actions. You stay in control of the decision.
How it works
I don't have access to your hiring system yet.
Add the Xobin connector and I can create assessments, screen candidates and build shortlists for you.
Add custom connector
MCP server
XobinOfficial
Hiring, assessments and talent intelligence
Xobin is successfully connected.
Claude can now work with your Xobin workspace. Revoke access any time.
Google Drive
Files and docs
Slack
Channels and messages
Allow Claude to use your Xobin workspace?
Signed in as john@yourcompany.com · Your Company
- Create and send skill assessments
- Read candidate profiles and scores
- Build shortlists and hiring reports
- Schedule interviews on your calendar
Xobin is now an official MCP connector in Claude.
Hiring, assessments and talent intelligence, natively available inside the assistant your team already uses.
Here is the same loop in a real assistant window, with the most common request first: finding people by what the evidence actually shows.
Illustrative example. Names, scores and evidence are fictional.
What you can do with it
1. Search and discoverability, first and most used
Keyword filters only find what you already knew to type. Xobin MCP lets a compatible assistant search supported profiles, résumés, assessment results and interview responses, and return the evidence behind each match.
- Find capabilities that never fit a filter: "who has organised a 300-person event?", "who has presented to a board?"
- Surface candidates already in your database before you spend on a new source.
- Ask about a whole pipeline, including strengths, gaps and where the shortlist is thin.
- Pull the report, score or transcript that supports any claim, in the same reply.
2. Compare and understand
Ask for a side-by-side on the two or three people who matter, framed by the requirements you care about, with trade-offs rather than a single ranking.
3. Create and distribute
Create job tracks, skills assessments, AI interviews and candidate forms from the requirements you just described, then send the invitations once you approve them.
4. Source and keep candidates moving
Send invites and messages, check interviewer availability, schedule live interviews, and set up stage changes for your approval.
5. Evaluate
Retrieve results as they arrive, review fit metrics and evidence, and turn what you learned into the next evaluation step.
Where it is available
Xobin MCP is a remote MCP server, so it works with any client that supports remote MCP connections over OAuth.
From a question to a reviewed result
You ask
A question in your assistant, in plain language.
Assistant calls Xobin MCP
The request becomes a specific Xobin tool call.
Xobin checks scope
Company, user role and approved permissions are applied first.
Evidence or action
Xobin returns the evidence, or prepares the action for approval.
You review and decide
It stays inside your Xobin permissions
Connecting an assistant does not widen anyone's access.
- You sign in through Xobin. Your password and Xobin API key are never shared with the assistant.
- You approve the scopes. The connector only gets the access you give it, and you can take that access away whenever you want.
- Xobin enforces context. Every request stays bound to the authenticated company, user and role permissions, following the same access boundaries as the Xobin app.
- Reads and writes are separate. Actions that change records or contact candidates are exposed as write operations, allowing compatible clients to apply their own approval or confirmation controls.
- You keep decision authority. The assistant can explain evidence and prepare actions; the hiring decision stays with your team.
That last point is deliberate. NIST's AI Risk Management Framework emphasizes explainability and clear human accountability when AI supports high-impact decisions. Hiring decisions clearly fall into this category.
Bring your hiring evidence into the assistant your team already uses.
Talk to the Xobin teamFrequently asked questions
- What is Xobin MCP?
- Xobin MCP is a standards-based connection that lets compatible AI clients request supported Xobin data and actions on behalf of an authenticated user.
- Who is it for?
- It is for recruiting teams that want to use Xobin context while working in an external AI assistant. Existing Xobin customers can bring the roles, candidates and hiring evidence already supported in their account into those conversations.
- Which AI clients can connect to Xobin MCP?
- Xobin MCP works with compatible clients like Claude, ChatGPT, and other AI apps that support remote MCP connections. Each client and plan may have different setup steps, available connectors, and approval requirements.
- What information can an assistant access?
- Access depends on the authenticated Xobin account, user role, approved scopes and supported MCP capabilities. A connected assistant does not receive unrestricted access to company or candidate data.
- Can an assistant change records or contact candidates?
- Supported actions can include creating recruiting resources, sending invitations or messages, scheduling interviews and preparing stage changes. Review the requested action before approving it, particularly when it affects a candidate or changes a hiring record.
- Is candidate data used to train the AI client?
- Data handling is governed by the AI client you choose and its applicable terms, settings and plan. Review that client's privacy and data-retention controls before connecting it to hiring data.
- Does Xobin MCP choose whom to hire?
- No. Xobin supplies available metrics and evidence, while the assistant helps retrieve and explain them. Recruiters remain responsible for defining job criteria, checking the supporting evidence and making hiring decisions.
- How do I get started?
- Contact the Xobin team to confirm availability for your account and get the setup details for your preferred MCP-compatible client.
Aman Kumar Tiwari
Aman writes about hiring practice, HR regulation and the assessment market. He tracks vendor releases and compliance changes so talent teams do not have to.
More from the Xobin hub
- NewsFive Things to Try First with Xobin MCPFive worked examples in a live assistant window, from evidence search to a context-built assessment, showing exactly how Xobin MCP works.
- NewsWhat we are measuring as recruiters start working with MCP agentsWe have begun tracking how recruiters use AI agents against live hiring data: the baseline metrics, where MCP fits the workflow, and the controls behind it.
- NewsWhy XobinA perspective on building trusted measurement infrastructure for talent in the AI era.