beevibe  by beevibe-ai

Agent-native OS for collaborative teams

Created 2 months ago
284 stars

Top 91.8% on SourcePulse

GitHubView on GitHub
Project Summary

Beevibe provides an agent-native operating system designed for companies, enabling human teams and AI agents to collaborate within a shared workspace. It addresses the challenge of fragmented AI work by creating a persistent, compounding intelligence layer, enhancing team coordination and context sharing beyond current human-centric tools.

How It Works

Beevibe employs a distributed architecture comprising a control plane (@beevibe/api) and a shared coordination/memory layer powered by Postgres. Local daemons (beevibe-daemon) run on user machines, spawning specific AI agent CLIs (e.g., Claude Code) within their local environments. This design allows agents to maintain persistent roles, develop bounded domain memory that deepens over time, and communicate via a "mesh" for inter-agent coordination, including asking specialists or escalating blockers to humans.

Quick Start & Requirements

  • Installation:
    • One-click deployment via Railway (requires API key configuration and CORS origin setup).
    • Local Docker Compose (docker-compose.quickstart.yml up -d --build) for a full stack setup.
  • Prerequisites:
    • API keys for services like Anthropic and OpenAI.
    • beevibe-daemon installation (e.g., brew install beevibe-ai/tap/beevibe-daemon).
    • Docker and Docker Compose for local deployment.
  • Links: See DEPLOYMENT.md for detailed deployment instructions and CONTRIBUTING.md for local development setup.

Highlighted Details

  • Agent-native OS facilitating collaborative human-AI teams.
  • Persistent AI agents with bounded domain memory that deepens across sessions.
  • Agent-to-agent coordination ("mesh") for requesting context or negotiating solutions.
  • Human-in-the-loop control via a web dashboard for oversight, redirection, and blocker resolution.
  • Self-hosted architecture where users retain ownership of their data and services.
  • "Bring Your Own CLI" model, allowing agents to run using user-configured tools and files on their local machines.

Maintenance & Community

Contributions are welcomed via Issues and Pull Requests. For significant changes, opening an issue first is recommended. Local development setup, DCO sign-off, and contribution license terms are detailed in CONTRIBUTING.md.

Licensing & Compatibility

The Beevibe source code is licensed under the Apache License 2.0. This license grants rights to use and modify the source code but explicitly reserves the "Beevibe" name and logo as project trademarks. Derivative works are permitted but must use distinct names.

Limitations & Caveats

Beevibe requires self-hosting and managing its infrastructure components. Users must install and configure the beevibe-daemon locally to enable agent execution. The system relies on external LLM APIs, necessitating user-provided API keys and incurring associated costs. The project appears to be in active development, with contributions encouraged for substantial feature additions.

Health Check
Last Commit

1 week ago

Responsiveness

Inactive

Pull Requests (30d)
11
Issues (30d)
1
Star History
132 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.