Guidebeginner
Getting Started with Google ADK: Build Your First AI Agent
Google's Agent Development Kit is an open-source framework for building AI agents. Here's what it does, why it matters, and how to run your first agent in under 10 minutes.
Open Agency guides
The blog for practical AI systems
Browse every published guide, filter by category and tag, and go deeper into practical AI workflows.
Filter by category
Filter by tag
Latest published guides
Guideintermediate
Configuring Oh My Opencode for Production: Model Routing, Cost Control, and Stack Integration
How to tune oh-my-opencode's per-agent model routing, manage concurrency limits, and integrate it with Context7 and OpenViking for production workflows.
Guide
Oh My Opencode Under the Hood: Architecture, Cost Modeling, and When Orchestration Is a Tax
An expert-level analysis of oh-my-opencode's four-layer architecture, per-agent model routing economics, and the engineering calculus behind multi-agent orchestration.
Guide
What Is Oh My Opencode and Why Should You Care?
A beginner-friendly look at the plugin that turns OpenCode into a multi-agent team, what it actually does, and whether it fits your workflow.
Guide
Token Optimization for AI Coding Agents: A Practical Guide to RTK
How RTK reduces token consumption by 60-90% through shell output compression, and how to integrate it into your agentic workflow alongside caching and dependency management.
Guidebeginner
How I Cut My AI Coding Costs by 80% With One Simple Tool
A beginner-friendly guide to RTK, the Rust CLI proxy that compresses shell output before it reaches your AI agent, saving 60-90% on token costs with zero workflow changes.
Guide
Running Kimi K2.6 on a Budget: What Actually Works
The open-weight release of Kimi K2.6 created a wave of expectations about cheap local inference. Most of those expectations are wrong. Here is what works, what does not, and how to make the right tradeoff for your stack.
Guide
Running Kimi K2.6 on a Budget: What Actually Works
The open-weight release of Kimi K2.6 created a wave of expectations about cheap local inference. Most of those expectations are wrong. Here is what works, what does not, and how to make the right tradeoff for your stack.