Kong (AI API Gateway) MCP. Manage APIs and AI traffic from your chat client.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Kong (AI API Gateway) MCP Server manages your entire API lifecycle. Your agent can list, create, and update services, routes, and plugins—including setting up secure LLM routing with `ai-proxy`.
It lets you configure rate limits, manage consumer keys, and audit the whole gateway without touching a UI. It’s full control over your backend APIs and AI traffic, all from chat.
What your AI agents can do
Create ai plugin
Applies a new plugin, like AI Proxy, to a specific service for LLM routing and key encapsulation.
Create consumer key
Generates a new API key credential for a specific Kong consumer profile.
Create route
Establishes a new routing rule, exposing an existing service via a defined path or hostname.
Create new backend services by defining the upstream URL, name, and protocol.
Set up inbound routing rules that map incoming client requests to specific backend services.
Apply, configure, and update plugins like ai-proxy to manage LLM routing and security features.
List consumer profiles and generate specific API keys for usage tracking.
List all plugins, services, and routes to check the current state of the API gateway.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Kong (AI API Gateway) MCP Server: 10 Tools for API Management
Use these 10 tools to provision, audit, and modify every aspect of your API Gateway, from defining backend services to controlling AI plugins.
019d75c2create ai plugin
Applies a new plugin, like AI Proxy, to a specific service for LLM routing and key encapsulation.
019d75c2create consumer key
Generates a new API key credential for a specific Kong consumer profile.
019d75c2create route
Establishes a new routing rule, exposing an existing service via a defined path or hostname.
019d75c2create service
Defines a new backend service by specifying its upstream URL, name, and protocol.
019d75c2delete plugin
Removes a plugin from the Kong Gateway, taking it out of service permanently.
019d75c2list consumers
Lists all consumer profiles registered within the Kong API Gateway.
019d75c2list plugins
Shows all plugins (like Rate Limiting, AI Proxy, Key Auth) configured globally or on specific services/routes.
019d75c2list routes
Retrieves a list of all currently configured routing rules in the API Gateway.
019d75c2list services
Lists every backend service registered and available within the Kong API Gateway.
019d75c2update plugin
Modifies the configuration of an existing plugin, such as adjusting rate limits or swapping AI model providers.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Kong (AI API Gateway), then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Kong API Gateway MCP Server - Manage Services & Routes
Your agent gives you total control over your API lifecycle. You'll list, create, and update services, routes, and plugins—including setting up secure LLM routing with ai-proxy. You can configure rate limits, manage consumer keys, and audit the whole gateway without ever touching a UI. It's full control over your backend APIs and AI traffic, all from chat.
Define Backend Services
You'll define a new backend service by specifying its upstream URL, name, and protocol using create_service. You can list every backend service registered and available in the gateway with list_services.
Configure API Routing
To set up inbound routing rules, you'll establish a new routing rule, mapping incoming client requests to specific backend services via create_route. You can retrieve a list of all currently configured routing rules in the API Gateway with list_routes.
Control AI Plugins
Your agent lets you apply a new plugin, like AI Proxy, to a specific service for LLM routing and key encapsulation using create_ai_plugin. You can modify the configuration of an existing plugin, such as adjusting rate limits or swapping AI model providers, with update_plugin. You can remove a plugin from the Kong Gateway permanently with delete_plugin and show all plugins configured globally or on specific services/routes using list_plugins.
Manage User Credentials
To generate a new API key credential for a specific Kong consumer profile, you'll use create_consumer_key. You can list all consumer profiles registered within the Kong API Gateway using list_consumers.
Audit Gateway Health
Check the current state of the API gateway by listing every backend service available with list_services, checking all configured routing rules with list_routes, and showing all active plugins with list_plugins.
How Kong (AI API Gateway) MCP Works
- 1 Subscribe to the server and provide your Kong Admin URL and Admin Token.
- 2 Your agent connects to the gateway, authenticating with the provided credentials.
- 3 You issue a command (e.g., 'Create a new service for the billing API') and the agent executes the necessary API calls.
The bottom line is, you manage your entire API infrastructure and AI traffic using conversational commands, without ever logging into the Kong Manager UI.
Who Is Kong (AI API Gateway) MCP For?
Platform Engineers and Backend Developers who hate switching context between terminals, UIs, and chat interfaces. You're the one who needs to provision a new service or adjust a rate limit right now without slowing down your flow. This is for the ops engineer who's tired of clicking through dashboards at 2 am.
Audits the global gateway configuration, lists all enabled plugins, and removes unused modules to maintain a clean proxy pipeline.
Creates new backend services and sets up routes for new endpoints without leaving their coding environment.
Monitors and adjusts AI Proxy settings, swapping LLM providers or adjusting rate limits based on cost and performance metrics.
What Changes When You Connect
- Provision New Endpoints: Use
create_serviceto define a new backend service with its upstream URL and protocol. Then, usecreate_routeto immediately expose it to clients, all without leaving your terminal. - Secure AI Integration: Apply and configure the
ai-proxyplugin usingcreate_ai_plugin. This lets you route LLM traffic and handle key encapsulation securely, making your APIs AI-ready. - Real-Time Operational Fixes: Need to change a rate limit or switch an AI model provider? The
update_plugintool lets you patch existing plugin configurations dynamically, keeping your service running smoothly. - Governance and Audit: Use
list_pluginsandlist_servicesto audit the entire gateway. You can find every active plugin or every registered service, ensuring nothing is forgotten. - User Management: Manage client access using
list_consumersandcreate_consumer_key. This allows you to track usage per tenant or user group, enforcing proper access control. - Clean Architecture: Run an infrastructure audit by calling
list_pluginsanddelete_pluginto identify and remove unused modules, keeping the proxy pipeline clean and efficient.
Real-World Use Cases
A new microservice needs an API endpoint.
The backend developer writes the code for the 'Inventory Service.' Instead of manually logging into Kong Manager, they tell their agent: 'Create a service for the inventory backend' (create_service), then 'Expose it at /v1/inventory' (create_route). The agent handles the whole deployment cycle, and the service is live immediately.
We need to upgrade our LLM provider for cost reasons.
The AI Ops Team Lead runs into unexpected high costs with OpenAI. They tell their agent to use update_plugin to modify the ai-proxy plugin settings. They switch the backend model provider from GPT-4 to Claude, adjusting the rate limits in the same chat session. The change takes effect instantly.
A new third-party partner needs API access.
The Platform Engineer needs to onboard a new partner. They first run list_consumers to check existing accounts, then use create_consumer_key to generate a unique API key. Finally, they might use create_ai_plugin to ensure the new key is properly authorized through the gateway.
Debugging a misrouted API call.
The developer suspects a service is using outdated credentials. They use list_services to verify the correct service name, then use list_plugins to check if the key-auth plugin is active on that specific route. This pinpoints the exact failure point without digging through logs.
The Tradeoffs
Manual UI Configuration
Spending 20 minutes navigating the Kong Manager UI, clicking through tabs, and copying/pasting URLs just to set up a simple new service route. It's slow, and you lose your place.
→
Instead, tell your agent: 'Create a new service for the billing API at billing.local' (create_service), and then 'Map it to the /billing route' (create_route). It takes two simple commands, and the job is done.
Over-relying on Documentation
Consulting the massive Kong API documentation just to figure out the correct JSON payload structure for updating a rate limit. It's complex and time-consuming.
→
Just tell your agent: 'Increase the rate limit on the user service by 50 requests per minute' (update_plugin). The agent handles the payload and the API calls for you.
Assuming Everything is Connected
Trying to modify a service or plugin without first checking if the required consumer exists. This leads to 404 errors and confusing troubleshooting.
→
Always start by running list_consumers to confirm your consumer base. Then, use create_consumer_key to generate the specific credentials you need before attempting any configuration changes.
When It Fits, When It Doesn't
Use this MCP Server if your job involves constant, high-stakes API infrastructure changes. Specifically, if you need to: 1) Define new upstream services and routes; 2) Manage complex AI/LLM plugins (like ai-proxy); or 3) Audit the entire gateway state (using list_plugins, list_services, etc.).
Don't use this if you just need to view a single API key or check a simple status. For those single checks, a dedicated CLI command might be faster. This tool is for the full lifecycle: creation, modification, and auditing. If you only need to list things, use the list_services tool first to scope your query.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Kong. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Setting up a new service shouldn't mean logging into a separate web UI.
Today, setting up a simple new API endpoint requires context switching. You write the code in your IDE, then you have to copy the base URL and protocol details into the Kong Manager UI. You click 'Add Service,' paste the details, then you repeat the process to create a new route that maps the client path to that service. It's tedious, and you lose your flow.
With this MCP server, you just tell your agent, 'Create a service for the billing API' (`create_service`), and then 'Route it to `/billing`' (`create_route`). The agent handles the whole sequence, making the entire process conversational. You stay in chat, and the service is live.
Manage API access and AI plugins with Kong (AI API Gateway) MCP Server
You used to have to manually check who had access, generate keys, and then manually apply the right plugins (like `ai-proxy`) to ensure the rate limiting was set correctly. This was a multi-step process prone to human error.
Now, you can run `list_consumers` to check user access, then `create_consumer_key` to issue a key, and finally `create_ai_plugin` to apply the necessary plugin—all in one conversation. The system handles the dependencies and the payload.
Common Questions About Kong (AI API Gateway) MCP
How do I use the `create_service` tool with Kong (AI API Gateway) MCP Server? +
The create_service tool requires you to provide the upstream URL, name, and protocol for the new backend service. You tell the agent these three pieces of data, and it registers the service in Kong.
Can I manage AI routing with the `create_ai_plugin` tool? +
Yes, create_ai_plugin is used to apply the ai-proxy plugin to a service. This allows you to configure LLM routing, select model providers, and manage key encapsulation for AI traffic.
What does the `list_plugins` tool show me? +
The list_plugins tool shows every plugin currently configured on your Kong Gateway, whether it's global or scoped to a specific service or route.
How do I generate a new API key using `create_consumer_key`? +
You run create_consumer_key and specify the consumer profile name. The agent then generates the required API key credential for that consumer in Kong.
Can I update rate limits using the `update_plugin` tool? +
Yes. update_plugin lets you adjust plugin settings in real time, like changing a rate limit or swapping out an AI model provider without downtime.
How do I use the `list_consumers` tool to check my registered API users? +
It shows all Consumer profiles registered in Kong. You can see who's connected and how many API keys they have, which is key for tracking usage.
What can I do with the `create_route` tool in Kong (AI API Gateway)? +
It creates a new Route definition. This lets you expose a backend Service using specific path or hostname rules, ensuring clients hit the right endpoint.
Can I delete a plugin using the `delete_plugin` tool? +
Yes, this permanently removes a Plugin from the Kong Gateway. Use it to clean up unused modules and keep your proxy pipeline simple.
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.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Langfuse (LLM Tracing & Evals)
Monitor LLM apps via Langfuse — track traces, manage prompt templates, and audit evaluation scores.
Relevance AI
Equip your AI to trigger custom autonomous agents, execute chained prompts, and manage unstructured knowledge datasets directly within your Relevance AI studio.
FlowiseAI
Build LLM orchestration flows visually with a drag-and-drop interface for creating AI chatbots, agents, and RAG pipelines.
You might also like
Shipmondo
Ship packages across Scandinavia and Europe with multi-carrier label printing, tracking, and return management from one platform.
BscScan (BNB Chain Explorer)
Access real-time BNB Chain data—check balances, track BEP-20/BEP-721 transfers, and inspect verified smart contracts directly from your AI agent.
Corescan (Core DAO Blockchain Explorer)
Explore the Core DAO blockchain — check balances, track transactions, inspect smart contracts, and monitor network stats directly via AI.