How MCP Connects Your ATS to AI Assistants
Marcus Johnson
Head of Engineering
What is MCP?
Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external tools and data sources. Think of it as USB for AI — a standardized way for Claude, and other AI assistants, to plug into your software.
For recruitment, this means your ATS isn't just a web app anymore. It's a data source that AI can query, reason about, and take action on.
What You Can Do
With Hirer.one's MCP integration, you can talk to Claude and say things like:
- "Show me all candidates in the interview stage for the Data Analyst role"
- "Who are our top 5 candidates by match score?"
- "Draft a rejection email for candidates who didn't pass the screening stage"
- "Schedule an interview with Sophia Mitchell for next Tuesday at 2pm"
- "What's our conversion rate from screening to interview?"
Claude queries your ATS data in real-time and responds with accurate, up-to-date information.
28 Tools, 8 Categories
Our MCP server exposes 28 tools organized into categories:
| Category | Examples |
|---|---|
| Candidate Management | Search, view profiles, move stages, add reviews |
| Job Management | List vacancies, pipeline overview, stage candidates |
| Scheduling | List interviews, schedule new ones, view calendar |
| Analytics | Hiring metrics, job performance, team activity |
| Communication | Email templates, draft personalized emails |
| Automations | List workflows, toggle automations |
Setting It Up
- Claude Desktop: Add the MCP server URL to your Claude Desktop config
- Claude.ai: Connect via the integrations panel (requires OAuth)
- Custom AI apps: Use the MCP SDK to build your own integrations
The MCP server runs on Hono (a lightweight HTTP framework) and serves data via HTTP/SSE transport. All tools return structured JSON that AI can reason about.
The Future of Recruiting
MCP represents a fundamental shift in how recruiters work. Instead of switching between tabs, copying data, and manually updating systems, you have a conversation with an AI that does the busywork for you.
The recruiter's job becomes what it should be: talking to people, making judgment calls, and building relationships. The data management happens in the background.