Complete guide to setting up HireSquire MCP with Claude Desktop. Get native access to candidate screening, status checking, and email generation directly in Claude.
Setting Up HireSquire MCP for Claude Desktop: Complete Integration Guide
The Model Context Protocol (MCP) gives Claude Desktop direct access to HireSquire's candidate screening capabilities. This guide walks you through setting up the MCP server so Claude can screen candidates, check status, and generate emails autonomously.
Claude can now:
• Create screening jobs
• Check job status
• Get candidate results
• Generate candidate emails
• Cancel running jobs
• Report hiring outcomes
What is MCP?
MCP (Model Context Protocol) is an open standard that lets AI assistants like Claude interact with external tools and services. Instead of using the REST API or CLI, Claude can call HireSquire directly through MCP - it's like giving Claude a native ability to screen candidates.
Prerequisites
Before starting, make sure you have:
- Claude Desktop installed on your machine
- HireSquire account (free tier works!)
- API token from your dashboard
- Node.js 18+ installed (for running the MCP server)
Step-by-Step Setup
Step 1: Get Your API Token
- Log in to your HireSquire dashboard
- Click Profile in the top navigation
- Scroll to API Tokens
- Click Create New Token
- Copy the token (you'll need it soon)
Step 2: Configure Claude Desktop
Open your Claude Desktop settings and add the MCP configuration:
- Open Claude Desktop
- Click Settings (gear icon)
- Scroll to Developer section
- Click Edit Config
- Add the following to your config:
{
"mcpServers": {
"hiresquire": {
"command": "npx",
"args": ["-y", "hiresquire-cli", "mcp"],
"env": {
"HIRESQUIRE_API_TOKEN": "YOUR_API_TOKEN_HERE",
"HIRESQUIRE_BASE_URL": "https://hiresquireai.com/api/v1"
}
}
}
}
Important: Replace YOUR_API_TOKEN_HERE with your actual API token. Keep this file secure - don't commit it to version control.
Step 3: Restart Claude Desktop
After saving the configuration, restart Claude Desktop to load the MCP server.
Step 4: Verify the Integration
Once restarted, Claude should have access to HireSquire tools. You can verify by asking Claude something like:
"What HireSquire tools do you have available?"
Claude should respond with a list of available tools.
Using HireSquire in Claude
Once configured, here are common ways to use HireSquire with Claude:
Creating a Screening Job
Simply describe the job and provide resumes:
"Screen these candidates for our senior developer position. The job description is: [paste description]. Here are the resumes: [paste resume content]"
Claude will use the create_screening tool to submit the job.
Checking Status
"What's the status of job #123?"
Getting Results
"Show me the top 5 candidates from job #123 with scores above 80"
Generating Emails
"Generate an interview invitation email for candidate #5 from job #123"
Available MCP Tools
When connected via MCP, Claude has access to these tools:
| Tool | Description |
|---|---|
create_screening |
Submit a new screening job |
get_screening_status |
Check job status and summary |
get_results |
Retrieve screening results with filters |
generate_email |
Generate interview invite, rejection, or follow-up emails |
create_interview |
Schedule interview and generate meeting link |
get_credit_balance |
Check your account balance and screening costs |
whoami |
Verify token and get profile information |
get_api_schema |
Get full OpenAPI specification for HireSquire |
test_webhook |
Test a webhook endpoint connectivity |
Example Workflow
Here's a complete example of using HireSquire with Claude:
User: I need to screen 10 candidates for a product manager role.
The job description:
"We're looking for a product manager with 3+ years experience,
experience with agile methodology, and strong communication skills."
The candidates:
[Candidate 1: 5 years PM experience, agile certified...]
[Candidate 2: 2 years PM experience, no certification...]
...etc
Claude: I'll screen these candidates for you using HireSquire.
[Uses create_screening tool with job description and resume content]
Great! I've submitted the screening job. It has 10 candidates and
should complete in about 2 minutes. Let me check the status...
[Uses get_screening_status]
The job is now complete! Here are the results:
🏆 Top Candidates:
1. Candidate 1 - Score: 94 (Strong match)
2. Candidate 3 - Score: 87 (Good match)
3. Candidate 7 - Score: 82 (Good match)
Would you like me to:
- Show details on any specific candidate?
- Generate interview invites for the top candidates?
- Filter by minimum score?
Troubleshooting
MCP Not Appearing in Claude
Problem: MCP server isn't showing up
Solutions:
- Verify Node.js 18+ is installed:
node --version - Check your config file has valid JSON (no trailing commas)
- Try restarting Claude Desktop completely
- Check the Claude logs for error messages
Authentication Errors
Problem: Getting 401 errors
Solutions:
- Verify your API token is correct in the config
- Check the token hasn't expired in your dashboard
- Make sure you're using the full token (not just the name)
Tool Not Available
Problem: Claude says tool isn't available
Solutions:
- Restart Claude Desktop to reload MCP server
- Check MCP is enabled in Settings → Developer
- Verify no errors in MCP server startup
Security Best Practices
- Never commit your config file - Add it to .gitignore
- Use environment variables instead of hardcoding tokens where possible
- Rotate tokens regularly - Create new tokens periodically
- Use the minimum required permissions - Don't give tokens more access than needed
Next Steps
Now that you have MCP set up:
- CLI Guide - For command-line automation
- REST API - For custom integrations
- Python SDK - For programmatic integration
- Webhooks - For event-driven workflows
MCP gives Claude native hiring superpowers. Combined with Claude's reasoning abilities, you now have an AI assistant that can understand your hiring needs, screen candidates intelligently, and help you make better hiring decisions.