# Kong (AI API Gateway) MCP for AI Agents AI Agent Connect

> Kong (AI API Gateway) lets you manage your API gateway through natural conversation. Orchestrate backend services, configure routes, and manage AI plugins like ai-proxy directly from your agent. It handles consumer keys and audits your gateway's plugin pipeline without you having to leave your chat interface.

## Overview
- **Category:** friends-mcp
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_v0JAMPJEymFJeA5BOdGXfYEBh1b3USJtcGdqANfK/ai-agent-connect
- **Tags:** api-gateway, microservices, traffic-management, service-orchestration, cloud-native, ai-ops

## Description

Managing a high-traffic API gateway usually involves a lot of context switching. You have to open a browser, log into a dashboard, find the right service, and manually type in URLs and protocols. It's a chore that pulls you away from actual development. This Connector changes that by letting you talk to your gateway instance like it's a teammate. You can spin up new backend services, map out routing rules, and swap out AI models on the fly just by asking. It handles the heavy lifting of configuration while you stay in your flow. If you're looking for a way to keep your infrastructure organized without the manual overhead, this is a solid addition to the Vinkius catalog. You get a direct line to your gateway's core functions, from managing consumer keys to auditing your active plugins, all through a single chat interface. Instead of hunting through menus to find a specific plugin configuration, you can just tell your agent what needs to change. It handles the technical details of the admin side while you focus on the high-level architecture of your services. It makes the entire lifecycle of your gateway accessible without needing to memorize every endpoint. You can quickly adjust rate limits to prevent overloading or swap out providers when a specific model hits a bottleneck. It turns a multi-step manual process into a simple conversation.

## Tools

### list_consumers
See every consumer profile currently registered in your Kong gateway. This helps you keep track of who is accessing your APIs.

### create_consumer_key
Generate a new API key for a specific consumer to track their usage. It simplifies how you manage access for different tenants.

### create_ai_plugin
Add the ai-proxy plugin to a service for secure LLM routing and key handling. This is the fastest way to enable AI traffic management.

### create_route
Expose a backend service by creating a new route with specific path rules. It lets you map client requests to your services instantly.

### create_service
Define a new backend service including its name, upstream URL, and protocol. This builds out your infrastructure via simple commands.

### delete_plugin
Permanently remove a plugin from your gateway to keep the pipeline clean. Use this to prune old modules during audits.

### list_plugins
View all active plugins like Rate Limiting or AI Proxy across your gateway. This gives you a clear picture of your current configuration.

### list_routes
Pull a full list of all routing rules currently active in the gateway. It helps you verify how traffic is being directed.

### update_plugin
Modify existing plugin settings to swap models or change rate limits instantly. It is perfect for handling sudden traffic spikes.

### list_services
Check which backend services are registered in your Kong instance. This is the quickest way to see your current service inventory.

## Prompt Examples

**Prompt:** 
```
List all services in my Kong gateway.
```

**Response:** 
```
I've found the following services registered in your gateway:

| Service Name | Upstream URL | Protocol |
| :--- | :--- | :--- |
| auth-api | https://auth.internal | https |
| inventory-service | https://inv.internal | grpc |
| ai-gateway-proxy | https://ai.internal | https |
| legacy-backend | https://old.internal | http |

Would you like to see the routing rules for any of these?
```

**Prompt:** 
```
Add the AI proxy to my auth-service using OpenAI.
```

**Response:** 
```
Applying the `ai-proxy` plugin to the `auth-service`... 

**Done.** 

I've configured the gateway to route traffic to OpenAI. Your gateway is now ready to handle AI requests for that endpoint securely.
```

**Prompt:** 
```
Who are our active consumers?
```

**Response:** 
```
I've identified 3 active consumers in your gateway:

*   **MobileApp_Prod** (2 active keys)
*   **ThirdParty_Partner** (1 active key)
*   **InternalAdmin_CLI** (1 active key)

Would you like me to generate a new key for a new partner?
```

## Capabilities

### Create backend services
Define upstream URLs and protocols for new services instantly.

### Map inbound routes
Link client requests to specific backend services based on paths.

### Configure AI proxies
Apply the ai-proxy plugin to handle LLM routing and keys.

### Update plugin settings
Adjust rate limits or swap models in real-time without restarts.

### Manage consumer keys
Generate and track API keys for specific users or tenants.

### Audit gateway plugins
Discover and remove unused modules to keep your proxy clean.

## Use Cases

### Rapid microservice deployment
A developer needs to spin up a new microservice. They ask their agent to create_service and create_route to make it live in seconds.

### Dynamic AI model swapping
An AI Ops lead sees high costs from one provider. They use update_plugin to swap the model provider in the ai-proxy config on the fly.

### On-demand partner access
A new partner needs access to the API. The admin uses create_consumer_key to give them a unique key for the gateway without manual entry.

### Gateway security auditing
A security audit is coming up. The engineer uses list_plugins to find and delete_plugin on any old, unused modules to clean the proxy pipeline.

## Benefits

- Stop switching tabs by using create_service and create_route to build infrastructure directly from your chat.
- Adjust traffic flow instantly with update_plugin to swap models or change rate limits under heavy load.
- Secure your AI traffic by using create_ai_plugin to handle LLM routing and key encapsulation in one step.
- Keep your gateway lean by using list_plugins and delete_plugin to audit and remove unused modules.
- Simplify user management by using list_consumers and create_consumer_key to handle API access quickly.

## How It Works

The bottom line is you get a conversational interface for your entire Kong API lifecycle.

1. Provide your Kong Admin URL and Token to your AI client.
2. Describe the service or route you want to create in plain English.
3. Receive a confirmation once the agent has updated your gateway configuration.

## Frequently Asked Questions

**Can the Kong (AI API Gateway) MCP help me manage LLM traffic?**
Yes. It allows you to configure AI proxy plugins, manage model providers, and handle key encapsulation through natural conversation with your agent.

**How do I use the Kong (AI API Gateway) MCP to create new routes?**
You can simply tell your agent to create a route by describing the inbound path and which backend service it should point to. It handles the configuration for you.

**Can I use the Kong (AI API Gateway) MCP to manage API keys?**
Yes, you can use it to list existing consumers and generate new API keys for specific users or tenants without leaving your chat interface.

**How does the Kong (AI API Gateway) MCP handle AI Proxy plugins?**
It gives you a direct way to apply and configure the ai-proxy plugin to your services, making it easy to manage LLM routing and security.

**Is the Kong (AI API Gateway) MCP good for auditing my gateway?**
It's excellent for audits. You can quickly list all active plugins and services to identify unused modules or verify your current configuration.

**Can I update rate limits with the Kong (AI API Gateway) MCP?**
Yes. You can use the Connector to update plugin configurations in real-time, which is perfect for adjusting rate limits during traffic spikes.

**Can I use this server to manage Kong's AI Proxy plugin?**
Absolutely. The `create_ai_plugin` tool is specifically designed to inject the `ai-proxy` plugin onto Services. You can define providers like OpenAI or Anthropic and manage model routing directly through your agent.

**How do I create a new API route through a conversation?**
Use the `create_route` tool and provide a JSON payload defining the paths and the Service ID it should point to. Your agent will handle the Admin API call to provision the routing rule instantly.

**Can my agent generate new API keys for consumers?**
Yes. The `create_consumer_key` tool allows you to provision new credentials for specific Consumers. This is perfect for onboarding new tenants or rotating keys for downstream applications securely.