This collection brings together the latest updates, releases, events, and insights from the Ortus ecosystem, covering BoxLang, ColdBox, and modern CFML development. From major product launches and AI advancements to in-depth technical guides and real-world modernization strategies, these resources highlight how developers and organizations are building scalable, future-ready applications on the JVM.
It also captures key moments from the community, including webinars and major events like Into the Box 2026, showcasing the ongoing innovation, collaboration, and evolution happening across the Ortus world.
Product Releases & Tools
-
Introducing skills.boxlang.io — The Open Agent Skills Ecosystem for BoxLang & the Ortus World
This post introduces skills.boxlang.io, a public, agent-agnostic registry of reusable AI “skills” for the entire Ortus ecosystem. It enables teams to define, version, and share structured AI knowledge across tools like Copilot and Claude, eliminating prompt duplication and creating a scalable, standardized way to power intelligent agents
-
Introducing skills.boxlang.io — The Open Agent Skills Ecosystem for BoxLang & the Ortus World
This post introduces skills.boxlang.io, a public, agent-agnostic registry of reusable AI “skills” for the entire Ortus ecosystem. It enables teams to define, version, and share structured AI knowledge across tools like Copilot and Claude, eliminating prompt duplication and creating a scalable, standardized way to power intelligent agents.
-
Introducing cbMCP — Your ColdBox App, Live to Every AI Agent
This post introduces cbMCP, a ColdBox module that turns your running application into a live MCP server for AI agents. It allows tools like Claude or Copilot to directly inspect routes, handlers, and system state in real time—eliminating guesswork and enabling smarter, context-aware AI development.
-
Build Cross-Platform Desktop Apps with BoxLang
This post introduces the new BoxLang Desktop Runtime, enabling developers to build cross-platform desktop apps using Electron, Vite, and BoxLang. It highlights a “write once, run anywhere” approach—allowing the same codebase to run on macOS, Windows, and Linux with minimal setup and no rewrites.
-
Introducing BoxLings! An Interactive Teacher for BoxLang and TDD/BDD
This post introduces BoxLings, an interactive CLI learning tool that teaches BoxLang through hands-on exercises and real test feedback. It combines 100+ progressive challenges with a TDD/BDD-first approach, helping developers learn by fixing code, reading tests, and building real skills step by step.
-
BoxLang Goes Serverless on Google Cloud
This post highlights how BoxLang expands its multi-runtime capabilities by enabling serverless deployments on Google Cloud. It shows how developers can run event-driven functions without managing infrastructure, taking advantage of automatic scaling, faster development cycles, and cost-efficient cloud execution.
-
ColdBox 8.1.0 Released: AI Routing, MCP, and BoxLang-First Power
This release introduces powerful AI-focused features like
toAi()andtoMCP()routing, enabling developers to instantly expose AI agents and MCP servers via REST endpoints. It also includes scheduler improvements, better cluster reliability, and deeper alignment with BoxLang as a first-class runtime.
Content & Resources
This article explores the isolation that often comes with CTO leadership and how it affects decision-making. It highlights strategies like building advisory circles and using structured frameworks to make more confident, high-impact decisions.
-
BoxLang AI Series: Complete Guide to Building AI Agents
This post brings together the full BoxLang AI series into a complete guide for building production-ready AI agents. It covers core concepts like tools, memory, and agent orchestration, showing how to design scalable, intelligent systems that can reason, act, and integrate with real-world data and APIs.
-
ColdFusion Modernization for UK Universities (Without Downtime)
This article explores how UK universities modernize legacy ColdFusion systems without disrupting critical services. It highlights strategies like phased upgrades, cloud adoption, and zero-downtime deployments to maintain uptime while improving performance, security, and scalability.
BoxLang Updates
-
BoxLang v1.13.0: Compatibility, Concurrency, and Formatter Maturity
This release focuses on stability and production readiness, delivering major improvements in CFML compatibility, concurrency handling, and runtime reliability. It also introduces a production-ready formatter with CI/CD support, along with security and performance fixes across the platform.
-
BoxLang AI v3 Has Landed: Multi-Agent Orchestration, Tooling, Skills and So Much More
This post announces the release of BoxLang AI v3, a major update that redefines how AI agents, models, and tools interact within the BoxLang ecosystem. It introduces a powerful AI Skills system for reusable, versioned knowledge, along with multi-agent orchestration, MCP server integration, and a revamped tooling architecture—enabling developers to build more scalable, modular, and intelligent AI-driven applications on the JVM.
Mini serie of BoxLang AI BoxLang AI Deep Dive
-
BoxLang AI Deep Dive — Part 1 of 7: The Skills Revolution (AI 3.0 Series)
This post kicks off a 7-part deep dive series on building production-ready AI systems with BoxLang AI 3.0. It introduces the concept of “AI Skills” as reusable, versioned knowledge modules that eliminate prompt duplication and inconsistency across agents. By treating instructions as structured, shareable assets, developers can create more scalable, maintainable, and consistent AI-driven systems across their applications.
-
BoxLang AI Deep Dive — Part 2 of 7: Building a Production-Grade AI Tool Ecosystem (AI 3.0 Series)
This post continues the BoxLang AI 3.0 deep dive series, focusing on how to build a scalable and production-ready AI tool ecosystem. It explores the internal architecture behind tools—introducing components like
BaseTool,ClosureTool, and the global tool registry—designed to handle lifecycle management, observability, and execution automatically. By abstracting complexity away from developers, BoxLang enables consistent, reusable, and modular tool integration across AI agents and workflows. -
This post explores how BoxLang AI 3.0 enables true multi-agent orchestration by introducing hierarchical agent structures where agents can delegate tasks to specialized sub-agents automatically. It highlights how agents are organized in a tree with built-in cycle detection, stateless execution, and per-call memory isolation—allowing developers to build scalable “AI teams” that collaborate efficiently without manual wiring or complex coordination logic.
-
This post dives into the middleware system in BoxLang AI 3.0, introducing a critical layer that sits between agent execution and actual LLM/tool interactions. It explains how middleware enables cross-cutting concerns like logging, retries, guardrails, and human-in-the-loop validation without modifying core agent logic. By using a hook-based lifecycle and composable middleware stack, developers can gain full control, observability, and testability of AI workflows—solving one of the biggest challenges in building reliable, production-grade AI systems.
-
This post explores the provider architecture behind BoxLang AI 3.0, focusing on how a single unified API can seamlessly support 17 different AI providers. It introduces a capability-based system that ensures type-safe interactions, prevents runtime errors, and allows developers to switch providers with zero code changes. By abstracting provider-specific logic into a structured hierarchy and transport layer, BoxLang eliminates vendor lock-in and enables flexible, future-proof AI integrations across cloud and local environments.
-
This post explores how BoxLang AI 3.0 implements advanced memory systems and Retrieval-Augmented Generation (RAG) to build AI applications that retain context and knowledge over time. It introduces two core memory types—standard memory for conversation history and vector memory for semantic retrieval—along with over 20 memory strategies, document loaders, and multi-tenant identity routing. By combining short-term context with long-term knowledge through hybrid memory, developers can build intelligent AI systems that are context-aware, scalable, and grounded in real data rather than relying solely on model responses.
-
BoxLang AI Deep Dive — Part 7 of 7: MCP — The Protocol That Connects Everything (AI 3.0 Series)
This final post in the BoxLang AI 3.0 series explores the Model Context Protocol (MCP), a standardized way for AI agents to discover and interact with tools across different systems and languages. It explains how BoxLang acts as both an MCP client and server—allowing developers to consume external tools or expose their own—while eliminating integration complexity. By adopting MCP, BoxLang enables a truly interoperable AI ecosystem where agents, tools, and services can seamlessly connect regardless of implementation.
-
How to Develop AI Agents Using BoxLang AI: A Practical Guide
This hands-on guide walks through how to build real-world AI agents using BoxLang AI, moving beyond simple chatbots into systems that can reason, act, and remember. It covers core concepts like tools, memory, and agents, and demonstrates how to create a production-ready support agent capable of querying data, calling APIs, and handling multi-step workflows. The article emphasizes a unified API approach, multi-provider flexibility, and scalable architecture for building intelligent, autonomous applications on the JVM.
Updates
-
BoxLang v1.12.0: Destructuring, Spread, Ranges, Watchers, Oh My!
This release introduces major language enhancements like destructuring, spread syntax, and a new range operator, along with real-time file watchers and performance improvements. It marks a shift toward more expressive, modern development features while continuing to improve stability and speed across the BoxLang ecosystem.
-
BoxLang AI v3.1 Released: Audio, Async, Parallel Pipelines, and More
This release expands BoxLang AI with audio capabilities, async execution, and parallel pipelines, enabling faster and more scalable AI workflows. It also adds new tooling, provider support, and stability improvements for production-ready AI applications.
Ortus Upcoming Events & Webinars
-
CFCamp Pre-Conference Workshops Led by Ortus Solutions
Ortus Solutions is hosting hands-on pre-conference workshops at CFCamp, designed to give developers practical experience with CFML, Ortus tools, and best practices before the main event kicks off.
Ortus Past Events
-
Join Our Webinar: Intro to BoxLang AI – One API to Rule Them All (Part II)
This post highlights a webinar held on April 16, 2026, focused on building production-ready AI workflows using BoxLang AI. It showcases how developers can leverage a single unified API to work across multiple AI providers, create autonomous agents, and build scalable RAG pipelines—demonstrating practical approaches to modern AI development on the JVM.
-
Into the Box 2026 is Here: Learn All the Details!
This post highlights Into the Box 2026, held from April 29 to May 1 in Washington, DC, under the theme “Modernization in Motion: Building a Dynamic Future.” The event featured deep dives into BoxLang, AI, cloud-native development, and modern architectures, along with hands-on workshops, networking opportunities, and full access to session recordings and resources.
Watch the Keynote 01 & 02
Join the Ortus Community
Be part of the movement shaping the future of web development. Stay connected and receive the latest updates on, product launches, tool updates, promo services and much more.
Subscribe to our newsletter for exclusive content.
Follow Us on Social media and don’t miss any news and updates:
Add Your Comment