Summary

micasa is a terminal-based TUI application designed to meticulously track home projects, maintenance schedules, appliances, and vendor quotes. It targets users seeking a private, local solution for managing household assets without cloud reliance, subscriptions, or account creation. The primary benefit is a centralized, keyboard-driven interface for comprehensive home management, enhanced by local AI-powered data querying.

How It Works

Built in pure Go with zero CGO, micasa leverages the Charmbracelet library for its TUI, GORM for data persistence, and SQLite for its backend. A core design principle is storing all user data, including attached files like manuals and invoices, within a single SQLite file. This approach simplifies backups and ensures data locality. A novel feature is its integration with local LLMs (via Ollama or OpenAI-compatible APIs), enabling users to query their home data using natural language. The TUI employs a modal, keyboard-driven navigation paradigm, inspired by VisiData, allowing for efficient data exploration and manipulation.

Quick Start & Requirements

• Installation: go install github.com/cpcloud/micasa/cmd/micasa@latest (requires Go 1.25+). Binaries for Linux, macOS, and Windows (amd64, arm64) are available on the releases page.
• Prerequisites: Go 1.25+ for direct installation. Local LLM integration requires a running Ollama server or an OpenAI-compatible API endpoint.
• Demo: Run micasa --demo to explore with sample data.
• Documentation: Full documentation is available at micasa.dev/docs.

Highlighted Details

• All data, including file attachments, stored in a single, easily backable SQLite file.
• Local LLM integration allows natural language querying of home data, translating questions into SQL.
• VisiData-inspired modal TUI with keyboard-driven navigation and column-level operations.
• Zero cloud dependency, no account creation, and no subscriptions required.

Maintenance & Community

The project is developed with AI coding agents and welcomes contributions, including AI-assisted pull requests after review. It utilizes a Nix development shell with pre-commit hooks for code quality. Specific community channels like Discord or Slack are not explicitly mentioned.

Licensing & Compatibility

Licensed under the Apache-2.0 license. This license is permissive and generally compatible with commercial use and linking within closed-source applications.

Limitations & Caveats

Direct installation via go install requires Go version 1.25 or newer. Setting up the local LLM integration necessitates a separate, pre-configured LLM server. The modal TUI may present a learning curve for users unfamiliar with such interfaces.