Glossary
The words used across Lyna and these docs, explained in plain language.
Terms are in alphabetical order. Each links to the page that explains it in full.
| Term | Meaning |
|---|---|
| Agent team | A group of sub-agents the AI assembles for a large job. They work in parallel and report back. See Agent teams. |
| Ask mode | A chat mode that answers questions without editing your project. See Chat modes. |
| Branch | A separate line of work in your project's Git history. See Git. |
| Build mode | The default chat mode. The AI makes changes directly. |
| Canvas | The area in Design view where your app's frames run and you edit elements. See Canvas. |
| Commit | A saved snapshot of your code in Git. |
| Composer | The message box at the bottom of the chat, with the mode menu, model picker and + menu. |
| Credits | The unit that measures AI usage on Lyna models on paid plans. See Credits per turn. |
| Customize | The settings page for plugins, MCP servers, subagents, rules, commands and hooks. |
| Effort | How much the model thinks before it answers, on models that support it. See Effort. |
| Frame | One view of your app on the canvas, at a given device size. |
| Hook | A command or instruction that runs at a set moment, such as after the AI edits a file. See Hooks. |
| Knowledge | Project guidelines the AI remembers across all edits. See Knowledge and memory. |
| Lyna Cloud | The built-in backend: database, users, storage, secrets, edge functions and payments. |
| MCP server | A service that gives the AI new tools through the Model Context Protocol. See MCP servers. |
| Memory | Facts the AI learns about you and your project and reuses later. |
| Mention | A file, folder, URL, Git changes or problems you add to a message with @. See Mentions and context. |
| Plan mode | A chat mode where the AI discusses and writes a plan before building. |
| Plugin | A package from the marketplace that adds skills, commands, subagents, MCP servers or hooks. |
| PRD | A product requirements board with tasks in columns. See PRD and tasks. |
| Provider | The service that serves a model: Lyna, Anthropic, OpenAI or Moonshot AI. See Models and providers. |
| Remix | Make your own copy of a template or a shared project. See Duplicate and remix. |
| Rule | An always-on instruction the AI follows in a project. See Rules. |
| Sandbox | The cloud machine that runs your project's code and development server while you build. |
| Skill | A reusable set of instructions the AI loads when a task matches it. See Skills. |
| Slash command | A command you run by typing / in the chat. See Slash commands. |
| Subagent | A focused helper the AI can hand a task to. See Subagents. |
| Template | A ready-made project you can remix. See Templates. |
| Version | A saved state of your project that you can preview and restore. See Version history. |
| Workspace | A group of projects and people with shared roles. See Workspaces. |