Discover and explore top open-source AI tools and projects—updated daily.
alvinunrealDesktop pets visualizing AI coding agent activity
New!
Top 38.6% on SourcePulse
Summary OpenPets provides a playful desktop companion for AI coding agents (e.g., Claude Code, OpenCode), visualizing agent progress, tool usage, and states via animations and speech bubbles. It targets developers seeking a more interactive and transparent experience with AI assistants, offering a delightful visual feedback loop that enhances engagement without compromising privacy.
How It Works
A tray-first desktop application (@open-pets/desktop) communicates with AI agents via local IPC. Agents integrate using the Meta Communication Protocol (MCP) via @open-pets/mcp or CLI tools (@open-pets/cli), or through specific hooks/plugins (@open-pets/claude, @open-pets/opencode). The desktop app interprets agent events (thinking, editing, testing, errors) to trigger corresponding pet reactions and static speech bubbles. This approach offers a novel, visual layer for agent status, prioritizing privacy by using local, static messages and avoiding exposure of prompts or code.
Quick Start & Requirements
Install the desktop app by downloading builds for macOS, Windows, or Linux from Releases. Connect agents via the desktop app's Integrations screen or by running npx -y @open-pets/cli configure --agent <agent> --pet <petId> from your project. Development requires Node.js 20+ and pnpm 11+. Documentation for integrations and testing is available in the docs/ directory.
Highlighted Details
Maintenance & Community The README provides no specific details on contributors, sponsorships, or community channels (e.g., Discord, Slack).
Licensing & Compatibility The repository's README does not explicitly state a software license. This omission requires clarification for assessing commercial use or closed-source linking compatibility.
Limitations & Caveats Current desktop builds may be unsigned, potentially triggering security warnings on macOS/Windows; a workaround is provided for macOS. The lack of a specified license is a critical factor for adoption decisions. Automatic hooks are best-effort and do not influence agent behavior.
3 days ago
Inactive