How to Set Up DocVerify with Claude Code: Complete MCP Integration Guide

What You'll Build

By the end of this tutorial, you'll have a Claude Code AI agent that can automatically detect fake receipts, forged PDFs, and manipulated documents before processing expense claims, loan applications, or onboarding documents.

Step 1: Get Your DocVerify API Key

First, you'll need a DocVerify API key to authenticate your requests.

1.1 Sign up for DocVerify

1. Go to DocVerify
2. Click "Sign Up" or "Get Started"
3. Complete the registration
4. You'll automatically get 10 free scans per month to test with

1.2 Get your API key

Once you're logged in:

1. Navigate to Admin Panel → API Keys
2. Click "Create New API Key"
3. Give it a name like "Claude Code Integration"
4. Copy the API key - you'll need it in Step 2

Step 2: Configure MCP in Claude Code

Claude Code uses the Model Context Protocol (MCP) to give AI agents native access to tools. We'll configure DocVerify as an MCP server.

2.1 Locate your Claude Code config

Claude Code stores MCP configuration in:

macOS/Linux: ~/.config/claude-code/mcp_settings.json

Windows: %APPDATA%\Claude Code\mcp_settings.json

2.2 Add DocVerify MCP server

Open mcp_settings.json and add the DocVerify server. The configuration should include the command, args, and your API key in the env section.

Replace your_api_key_here with your actual API key from Step 1.

2.3 Restart Claude Code

Restart Claude Code to load the new MCP configuration:

• macOS: Cmd + Q then reopen
• Windows/Linux: Close and reopen the application

Step 3: Verify the Integration

Let's test that DocVerify is properly connected.

3.1 Check available tools

In Claude Code, open a new chat and ask:

"What MCP tools do you have access to?"

You should see DocVerify listed among the available tools with analyze and batch operations.

3.2 Test document verification

Create a simple test by asking Claude Code:

"Can you check if this receipt is authentic?"

Then upload a receipt image or provide a URL.

Claude Code will automatically call the docverify_analyze tool and return results showing the authenticity score, suspicious flags, and verdict.

If you see this response, you're all set! 🎉

Real-World Use Cases

Now that you have DocVerify integrated, here are practical ways to use it:

Use Case 1: Automated Expense Review

Problem: Employees submit fake or manipulated receipts to inflate expense claims.

Solution: Ask Claude Code to review expense claims and flag suspicious receipts. It will automatically verify all receipts using DocVerify and flag any that show signs of manipulation.

Use Case 2: Loan Application Verification

Problem: Applicants submit forged bank statements or pay stubs to qualify for loans.

Solution: Have Claude Code verify all loan application documents and provide a comprehensive authenticity report.

Use Case 3: HR Onboarding Document Verification

Problem: New hires submit fake employment letters or proof of income for benefits eligibility.

Solution: Ask Claude Code to review onboarding documents for authenticity and flag any documents that appear forged or manipulated.

Troubleshooting

"MCP server not found"

Problem: Claude Code can't find the DocVerify MCP server.

Solution:

1. Verify mcp_settings.json is in the correct location
2. Check that npx is installed by running: npx --version
3. Restart Claude Code completely

"Authentication failed"

Problem: Invalid or missing API key.

Solution:

1. Verify your API key in mcp_settings.json
2. Check that you copied the full key (no trailing spaces)
3. Generate a new API key if needed from the admin panel

What's Next?

Now that you have DocVerify integrated with Claude Code, you can:

• Build custom workflows for your specific fraud detection needs
• Set up automated screening for expense claims, loan applications, or onboarding
• Create agent teams that combine document verification with other checks
• Monitor fraud patterns by tracking suspicious documents over time