Wire your LLM into the agent
● MCP endpoint · drop-in config for Claude / Cursor / OpenAI
MCP endpoint
https://pay-on-arc.vercel.app/mcp Server launching with Vercel deploy · stub for now
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (Mac) or %APPDATA%/Claude/claude_desktop_config.json (Windows):
{
"mcpServers": {
"arc-agent": {
"url": "https://pay-on-arc.vercel.app/mcp",
"transport": "sse"
}
}
}Cursor
In Cursor Settings → MCP, add a new server:
{
"mcpServers": {
"arc-agent": {
"url": "https://pay-on-arc.vercel.app/mcp"
}
}
}OpenAI / generic function-calling
Or expose the agent directly as a tool to any LLM with function calling:
{
"type": "function",
"function": {
"name": "pay",
"description": "Send USDC from the agent's wallet to a recipient on Arc.",
"parameters": {
"type": "object",
"properties": {
"to": { "type": "string", "description": "Recipient address (0x…)" },
"amount": { "type": "number", "description": "USDC amount (max 1.00)" },
"memo": { "type": "string", "description": "Optional reason" }
},
"required": ["to", "amount"]
}
}
}What the agent will accept
Any payment instruction within the rules on the Constitution. Outside the rules, the agent rejects with a structured error the LLM can read and try again with adjusted parameters.