TuxNest
Download
TuxNest/Docs/Getting started

Documentation

Getting started

Install TuxNest, verify the active extension, connect an existing model runtime and run a deliberately small first task.

Requirements

  • VS Code 1.106 or later.
  • A trusted workspace for normal use; a disposable workspace is recommended for the first Agent task.
  • An existing local Ollama installation, a user-owned remote Ollama host, or an explicit compatible API endpoint.

Install the extension

  1. Download the verified VSIX you intend to run.
  2. In VS Code, open the Command Palette.
  3. Run Extensions: Install from VSIX….
  4. Select the package and reload the window.
Verify what actually loaded.Run TuxNest: Diagnose Installation after installation. A filename alone is not proof of the active extension version.

Open TuxNest

Use TuxNest: Open TuxNest SI Agent or open TuxNest Chat in the Secondary Sidebar. Model and mode selectors live near the composer.

Select a real model

Open Models and refresh discovery. Local Ollama models are discovered from the runtime; TuxNest should not invent a model that is not actually available. For remote GPU use, connect the host first.

First task

Start with a bounded read-only request:

Read README.md and package.json.
Summarize the actual project and commands.
Separate inspected facts from assumptions.

Only increase scope after reads, approvals, edits and verification behave as expected with the selected model.

← DocumentationModels & providers →