A practical guide to OpenCode concepts. What each thing is, why it exists, and how to use it.
A specialized AI assistant configured for specific tasks and workflows.
A specialized assistant invoked by primary agents for narrow tasks.
Reusable instruction definitions loaded on demand by agents.
Functions the LLM can call to take actions in your codebase.
Custom slash commands for repetitive tasks in the TUI.
Model Context Protocol servers that add external tools and services.
Event-driven callbacks that let plugins react to OpenCode events.
JavaScript/TypeScript modules that extend OpenCode's behavior.
A conversation context between you and an OpenCode agent.
The information visible to the LLM during a session.
Each page explains what something is, why it exists, where it lives in your project, how to create it, and includes a minimal copyable example. Content is grounded in real OpenCode documentation.
Start with Agent if you're new, or browse the full concept index.