vmblu is a toolchain for building and maintaining software with coding agents. It includes schemas, skills, a CLI, a runtime and a graphical editor. At its core is an executable blueprint: a structured, machine-readable model of nodes, routes and contracts.

The blueprint is not documentation beside the system. It is the system architecture, the authoritative source for generation, testing, runtime behavior and agent interaction.

vmblu works with your existing stack. Keep your frameworks, libraries and language choices. Today it ships as a VS Code extension, with other environments to follow.

vmblu models the architecture of a system as a collection of interacting nodes. Nodes communicate through pins with formal message contracts. Complex nodes can themselves be modeled as groups of nodes.

Functionality is localised and data flow is visible. Humans and agents get a compact architectural context that reads like a circuit diagram. A vmblu blueprint is not a picture of the system, it is the system.

That modularity matters for coding agents. An agent can work on one part of the system without loading the entire codebase into context. Changes stay local, dependencies stay explicit, and mistakes have a smaller blast radius.

A vmblu system exposes explicit capabilities that can be used by a browser, a mobile app, another service or directly by an agent.

The architecture defines what those capabilities are. The runtime controls who can use them and what they are allowed to do.

Build software as an app, a service, an agent capability or all of them, from the same architecture.

Coding agents can generate enormous amounts of code. You should not have to read all of it to understand your system.

vmblu keeps the architecture explicit. You can see what every node does, how data flows through the system and which contracts connect its parts.

Generated code is profiled against the model and interfaces are checked against their contracts. Behavior stays localized, making failures easier for both humans and agents to trace.

Agents can generate node-level tests and reports against the architecture. When you do need to inspect the code, modular boundaries make that inspection manageable.

Understand the system without reading the whole system.

vmblu includes a runtime for message switching between nodes. That runtime also acts as an enforcement layer: it applies security policies and exposes controlled capabilities to agents.

Security: nodes can be restricted in what they are allowed to access (network, filesystem, processes, etc.). If a node exceeds its permissions, the action is blocked and logged. Trust cannot rely on static analysis alone; vmblu enforces policy while the system runs.

Capabilities: external agents can interface with a vmblu application at a fine-grained level using mcp or other protocols. Capabilities allow an agent to do what it has to do, ...and nothing more.

vmblu is released under the Apache-2.0 license—permissive and business-friendly.

Use it in commercial products, modify it freely, and contribute back to a growing ecosystem.