HN · GitHub Signals

OPEN SOURCE, IN FOCUS

osaurus

osaurus-ai/osaurus

Own your AI. The native macOS harness for AI agents -- any model, persistent memory, autonomous execution, cryptographic identity. Built in Swift. Fully offline. Open source.

Explore on GitHub ↗

Saved projects stay in this browser. No account needed.

TAKE A CLOSER LOOK

Start with the source.

The repository is the place to check installation steps, supported platforms, the license, and current maintenance before choosing a tool.

Read the documentation on GitHub ↗

Keep exploring

Find projects in the same areas, or search the archive for a different approach.

ANOTHER GOOD FIND

Explore related projects

View the archive ↗

Connected by topics and descriptions.

macos26

agent

Agent! Is a native macOS App that works with any LLM [local and cloud], [open source and paid] Claude, OpenAI, Gemini, Ollama, Qwen, Hugging Face, and 12 more. Writes code, builds Xcode projects, bump versions, manages git, automates Safari (AppleScript, JS, or Accessibility), extends via MCP servers. Accepts tasks from your iPhone via Messages. 🤯

Take a look →
mihaiciprianchezan

agentic-first-enterprises

A reference operating model for organizations where any role can be filled by a human or an agent, every process is built for agents first, and every flow can be paused, adjusted by a human, and resumed.

Take a look →
withkynam

vibecode-pro-max-kit

Your AI forgets. This remembers. Spec-driven coding harness for vibecoders, product owners, CEOs and real builders — self-improving context memory, 15 agents, 33 skills working with /goal, agent-team, & workflow on autopilot loops with 0 need for human gate. Kills context rot, ships features, not spaghetti. Claude Code & Codex. Any stack

Take a look →
amap-ml

longhorizon-harness

The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.

Take a look →