Skip to content
Updated Jul 18, 2026

Connect An Agent

Connect your AI agents to AgentPMT to enable tool and workflow execution. AgentPMT supports two primary connection methods: Model Context Protocol (MCP) for modern AI platforms, and REST API for custom integrations.

MCP Integration

The Model Context Protocol (MCP) provides a standardized way for AI agents to discover and use tools from AgentPMT. This is the recommended method for connecting AI agents like Claude, ChatGPT, and other MCP-compatible platforms.

30 Second Setup

STDIO connector for Claude Code, Codex, Cursor, Zed, and other LLMs that require STDIO or custom connections.

bash
npm install -g @agentpmt/mcp-router
agentpmt-setup

Hosted Streamable HTTPS

MCP endpoint for browser-based apps like ChatGPT, Claude, Grok, or any time you want a streamable connection with no local install.

text
https://api.agentpmt.com/mcp

Config Example

Use the hosted endpoint directly in clients that support remote MCP. Store your Bearer token in the client config or secret field.

json
{
  "mcpServers": {
    "agentpmt": {
      "type": "streamable-http",
      "url": "https://api.agentpmt.com/mcp",
      "headers": {
        "Authorization": "Bearer <AGENTPMT_BEARER_TOKEN>",
        "x-instance-metadata": "{\"client\":\"generic-mcp\",\"platform\":\"remote\"}"
      }
    }
  }
}

For platform-specific setup and the full connection guide, see Connection.

RestAPI Integration

For custom integrations and advanced use cases, AgentPMT provides a REST API that gives you programmatic control over tool execution, workflow management, and agent interactions.

API Overview

The REST API allows you to:

  • Execute tools directly via HTTP requests
  • Manage workflows and their execution
  • Handle authentication programmatically
  • Monitor usage and billing
  • Integrate with existing systems using standard web protocols

API Reference

For complete API documentation, see the Programmatic Access reference.