MCP Server
Enable AI assistants to generate invoices with the Model Context Protocol.
What is MCP?
The Model Context Protocol (MCP) is an open standard for connecting AI assistants to external tools and data sources. Our MCP server allows Claude, GPT, and other AI assistants to generate and validate invoices.
Installation
Claude Code
claude mcp add @envoice/mcp-server
Manual Configuration
Add to your MCP settings file:
{
"mcpServers": {
"envoice": {
"command": "npx",
"args": ["@envoice/mcp-server"]
}
}
}
Available Tools
envoice_generate
Generate a ZUGFeRD-compliant PDF invoice from structured data.
| Parameter | Type | Required |
|---|---|---|
api_key |
string | Yes |
template |
string | No |
locale |
string | No |
invoice |
object | Yes |
envoice_validate
Validate an existing PDF for ZUGFeRD/Factur-X compliance.
| Parameter | Type | Required |
|---|---|---|
api_key |
string | Yes |
pdf_base64 |
string | Yes |
Example Prompt
"Generate an invoice for 10 hours of consulting at 150/hour from Acme GmbH (Berlin) to Customer AG (Munich), due in 14 days. Use my API key: env_sandbox_xxx"
Direct HTTP Access
You can also access MCP tools via HTTP without the npm package.
Our API exposes a JSON-RPC 2.0 endpoint at /mcp:
# List available tools
curl -X POST https://api.envoice.dev/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
# Call a tool
curl -X POST https://api.envoice.dev/mcp \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/call",
"params": {
"name": "envoice_generate",
"arguments": {
"api_key": "env_sandbox_xxx",
"invoice": { ... }
}
}
}'