Skip to main content

Tutorials

You don't need to understand every Helix feature before you start. This series follows the real first-run path: connect Helix to a project, learn the main interface, configure a model, start a conversation, then use Manager mode, terminal, file browser, and settings when you need them.

Follow the order

If you're new to Helix, start with workspace setup. The first two tutorials establish where Helix can work; the rest of the series builds on that real project context.


📖 Onboarding Path

1. Connect a Workspace and Machine

Connect a local folder, remote server, or VM workspace first. After this step, Helix knows where it can read files, run commands, and help you edit code.

2. Get to Know the Helix Interface

Learn the workspace area, session list, Chat / Manager switch, tool buttons, and bottom model bar so you can find the right control when you need it.

3. Configure Your First Model

Use built-in Helix account access, or connect your own DeepSeek, OpenAI, Anthropic, or OpenAI-compatible provider and set a default model.

4. Start Your First Conversation

Create a session, send a useful first prompt, watch the AI read your project and call tools, then learn how to interrupt, follow up, and switch models.

5. Use Manager Mode

For tasks that span multiple files or steps, hand the goal to Manager so it can plan the work, run sub-tasks in parallel, and summarize the result.

6. Terminal & File Browser

Open the built-in terminal, browse the project tree, preview code, and combine manual checks with AI-driven changes in the same workspace.

7. Settings & Customization

Find the settings that matter: LLM providers, MCP tools, Skills, appearance, language, and advanced preferences.


Ready? Start with Connect a Workspace and Machine. By the end of that tutorial, you'll have a Helix workspace ready for chat, commands, and code changes.