The Model Context Protocol (MCP) is rapidly becoming the gold standard for connecting Large Language Models to external tools and data sources. Historically, if you wanted to build an MCP client or server, you needed to reach for heavy, bloated SDKs in Node.js or Python, dealing with complex asynchronous state and dependency hell.

We thought: What if we could do it entirely natively in Coni?

In this post, we’re going to walk through how we brought a complete, zero-dependency MCP Client and Server directly into the Coni standard library (mcp.coni). It’s fast, highly concurrent, and ridiculously easy to use.

Let’s face it: writing 2D browser games from scratch can be a total drag. Between managing Canvas contexts, handling asynchronous asset loading, and dealing with Javascript’s eccentricities, you often spend more time wrestling with boilerplate than actually designing your game.

Enter Namakemono—our incredibly cute and dangerously fast 2D game framework built natively in Coni targeting WebAssembly.

The Single State Philosophy

One of the biggest headaches in game development is state management. Do you keep your player, enemies, and projectiles in separate variables? What happens when an enemy needs to check the player’s position?

In our ongoing quest to make Coni the most capable, dependency-free, and relentlessly fun functional language out there, we’ve hit a massive milestone. We didn’t just write an AI script—we went fully native.

Today, we’re unveiling our first fully native AI example: The Poem Generator.

Using the highly capable Qwen 2.5 0.5b instruct model formatted as a raw GGUF binary, this new Coni script takes a theme, invokes a local LLM, computes a beautiful poem, and spits out characters at a shockingly fast rate of over 70 tokens per second (TPS).

Building a multi-agent orchestration system from scratch often means dealing with endless layers of complexity: orchestrating interpreters, managing external tool executions, and wrangling the rigid schema constraints of external LLM APIs.

But Coni is the answer to complexity.

With Coni Agent Studio, we’ve built a streamlined, native orchestration layer where agents don’t just chat—they act, execute, and solve real-world engineering problems autonomously.

To demonstrate the power of this system, we built an end-to-end swarm dedicated to one of the most tedious tasks in software engineering: Javascript Dependency Security and CVE Patching.

In the last 24 hours alone, we’ve pushed the boundaries of the Coni WebAssembly ecosystem to produce an entire suite of procedural audiovisual web experiences. By combining the Coni macro system with raw WebGL and the WebAudio API, we successfully scaffolded, compiled, and deployed complex graphical and musical systems natively in the browser—with absolutely zero external asset dependencies.

Let’s take a look at the five new applications we’ve built. You can experience them all natively in your browser on the Coni WASM Gallery.

Mental health and focus are paramount in today’s fast-paced digital world. Following the success of our brainwave entrainment applications, we decided to tackle another facet of well-being: conscious breathing.

We are excited to unveil our latest Coni WASM application: the WebGL Breathing App, a visual and auditory experience built from scratch to guide you through a calming 6-2-6 breathing exercise (6 seconds inhale, 2 seconds hold, 6 seconds exhale).

Creating a Living UI

Unlike traditional breathing timers that simply show a growing circle, we wanted to build something that feels alive. We designed an interactive menu where you can set your Mood—choosing from Calm (Cyan), Warm (Orange), or Focus (Purple).