Tolaria

Overview

Tolaria is a desktop app for macOS, Windows, and Linux for managing Markdown knowledge bases. It is files-first, Git-first, and offline-first: notes are plain Markdown files with YAML frontmatter, each vault is a Git repository, and no account or cloud dependency is required. The app combines a block-based editor, wikilinks, relationships, a built-in Git client, and AI support through CLI coding agents or local/API model providers.

Key Features

• Plain Markdown files on disk with YAML frontmatter, no database or proprietary format.
• Block-based editor with slash commands, wikilinks, raw Markdown, whiteboards, media previews, table navigation, and note width controls.
• Fully integrated Git client for committing, pushing, browsing history, and viewing per-note version history.
• AI support with CLI coding agents such as Claude Code, Codex, OpenCode, Pi, and Gemini, plus local or API model providers for chat over note context.
• Native relationships and types as lenses, not schemas, for navigating notes.
• Cross-platform desktop app for macOS, Windows, and Linux with offline-first operation and no account required.

Use Cases

• Operate second brains and personal knowledge.
• Organize company docs as context for AI.
• Store OpenClaw/assistants memory and procedures.
• Manage large personal vaults for journaling, writing, and knowledge management.

Getting Started

• On macOS, install via Homebrew with brew install --cask tolaria.
• Download the latest release for macOS, Windows, or Linux.
• On first launch, you can clone the Getting Started vault for a walkthrough of the app.
• Public user docs start with Install Tolaria and First Launch.

Deployment & Requirements

• Desktop app supports macOS, Windows, and Linux.
• For local development: Node.js 20+, pnpm 8+, Rust stable, and macOS or Linux.
• Linux development requires WebKit2GTK 4.1 and GTK 3, with distro-specific packages.
• The bundled MCP server spawns the system node binary at runtime on Linux for the external AI tooling flow.
• Windows installers are Authenticode-signed; company-managed devices may still require IT approval of the Tolaria publisher before first install.

Before You Adopt

• License: AGPL-3.0. Review its terms before using, modifying, or distributing the project.
• Windows installers are signed, but company-managed devices may still require IT approval before first installation.
• The external AI tooling flow on Linux depends on the system node binary for the bundled MCP server.
• Tool-backed AI editing depends on external CLI coding agents when that workflow is used.
• Chat over note context can rely on local or API model providers and does not provide vault-write tools.