This started as an internal experiment to explore how AI could assist with API design, using similar standards to what Kong R&D applies to our own API design. It turned out to be useful enough that we decided to share it more broadly.
Now it’s live at [create-api.dev](https://create-api.dev)create-api.dev. No login. No setup. Just describe an API and get a structured OpenAPI spec you can refine, share, and eventually ship.
Many popular OpenAPI design tools are either dated or too heavyweight for early-stage API design. We wanted something faster. Minimal UI. Simple collaboration. Type a prompt or upload a design or spec, and create-api.dev will generate an OpenAPI spec. Share the link. That’s it.
Rendering your designed spec in create-api.dev was also a chance to show off our newly open sourced spec renderer, the same one used in the [Konnect Dev Portal](https://konghq.com/products/kong-konnect/features/developer-portal)Konnect Dev Portal (which you can try for free). There are plenty of spec rendering tools out there, but our team put serious effort into building something lightweight, fast, and customizable. [Check out the project](https://github.com/Kong/spec-renderer)Check out the project.
As you may imagine, the system prompt we use includes some basic instructions for generating an OpenAPI spec — but we also have incorporated design guidelines we use inside Kong R&D (see it [here](https://konghq.com/blog/engineering/appendix-system-prompt-snippet)here). As far as models, we’re currently using Google’s Gemini because the large context window works well for OpenAPI specs.
For small to mid-size REST APIs, create-api.dev has produced solid results in our internal testing. We're also tracking spec quality with an eval system (we're starting with [DeepEval](https://github.com/confident-ai/deepeval)DeepEval) as we react to feedback and continue to update the system prompt
We welcome your feedback to help improve! After you generate a spec, you can send feedback directly in the app. We use that input as part of our ongoing improvement process.
Here are three common starting points we've seen:
Once you design a spec, you can:
We don’t help you build the backend yet, but this shows how spec-first design fits into a modern API workflow and how Kong tools can support that process.
We’re excited to see what you build.
👉 [Try it at create-api.dev](https://create-api.dev)Try it at create-api.dev
We’ll share more details in a future post, but create-api.dev isn’t just a standalone tool. It’s also a working demo of how teams of any size can ship better AI-powered apps using Kong’s production-grade infra products.
All model requests go through Kong’s AI Gateway, the same system many of our customers use for their own AI applications. It gave us a faster way to add basic protections, routing, and visibility into how prompts are handled across different models and environments.
We're using Gemini in this setup, but AI Gateway makes it easy to route and A/B test across other model providers as well.
One example of what AI Gateway makes possible: we’re using the Semantic Prompt Guard plugin to block prompt injection attempts before they reach the model. Even this simple setup (a short deny list) was easy to implement and effective at blocking common attacks. It also helps reduce unnecessary token usage.
"deny_prompts": [
"system prompt",
"Ignore previous instructions"
]Routing, observability, and prompt filters like this are built in. If you want to learn more, you can [get started](https://konghq.com/products/kong-konnect/register)get started with AI Gateway in Konnect or [talk to our technical sales team](https://konghq.com/contact-sales)talk to our technical sales team.
APIs have quietly powered the global shift to an interconnected economy. They’ve served as the data exchange highways behind the seamless experiences we now take for granted — booking a ride, paying a vendor, sending a message, syncing financial rec
[](https://konghq.com/blog/engineering/api-ecosystems-for-the-agentic-era)
In the latest Kong Gateway 3.12 release , announced October 2025, specific MCP capabilities have been released: AI MCP Proxy plugin: it works as a protocol bridge, translating between MCP and HTTP so that MCP-compatible clients can either call exi
[](https://konghq.com/blog/engineering/ai-gateway-mcp-gateway-mcp-server-breakdown)
Bring Financial Accountability to Enterprise LLM Usage with Konnect Metering and Billing Showback and chargeback are not the same thing. Most organizations conflate these two concepts, and that conflation delays action. Understanding the LLM showb
[](https://konghq.com/blog/enterprise/llm-cost-management-ai-showback-and-chargeback)
Kong Gateway is an API gateway and a core component of the Kong Konnect platform . Built on a plugin-based extensibility model, it centralizes essential functions such as proxying, routing, load balancing, and health checking, efficiently manag
[](https://konghq.com/blog/engineering/ai-voice-agents-kong-ai-gateway-cerebras)
AI observability extends traditional monitoring by adding behavioral telemetry for quality, safety, and cost metrics alongside standard logs, metrics, and traces Time-to-First-Token (TTFT) and token usage metrics are critical performance indicator
[](https://konghq.com/blog/learning-center/guide-to-ai-observability)
Why AI guardrails matter It's natural to consider the necessity of guardrails for your sophisticated AI implementations. The truth is, much like any powerful technology, AI requires a set of protective measures to ensure its reliability and integrit
[](https://konghq.com/blog/engineering/ai-guardrails)
Introduction to OWASP Top 10 for LLM Applications 2025 The OWASP Top 10 for LLM Applications 2025 represents a significant evolution in AI security guidance, reflecting the rapid maturation of enterprise AI deployments over the past year. The key up
[](https://konghq.com/blog/engineering/owasp-top-10-ai-and-llm-guide)