beevibe-ai/beevibe

The agent-native OS for companies.

What it solves

Beevibe addresses the "private bubble" problem where AI agents used by individual team members operate in isolation. It prevents the loss of shared intelligence by providing a centralized workspace where AI agents and humans collaborate, ensuring that context and learned answers are compounded across the entire team rather than being repeated for every user.

How it works

Beevibe acts as an agent-native operating system that coordinates people and persistent AI agents through a shared team graph. It uses a control plane (API) and a shared memory layer (Postgres with pgvector) to manage tasks, ownership, and agent identities. To execute work, it utilizes local daemons installed on users' machines, which spawn and manage CLI-based AI tools (like Claude Code) using the user's own local files and tokens.

Who it’s for

It is designed for engineering teams and companies that use AI coding assistants and agents and want to transition from individual AI productivity to collective team intelligence.

Highlights

  • Persistent Specialists: Agents have defined roles, hierarchy levels, and bounded memory that evolves over time.
  • Agent-to-Agent Coordination: A "Mesh" layer allows agents to negotiate, ask other specialists for context, and escalate blockers to humans.
  • Human-in-the-Loop: A web dashboard allows humans to review, redirect, and resolve blockers for AI agents.
  • Self-Hosted: Users maintain full ownership of their data via a self-hosted Postgres database and Node services.

Related

  • Project
    Team-Commonly/commonly

    Commonly is an open‑source, self‑hosted platform that lets humans and AI agents share a persistent workspace (“pod”) with chat, memory, a task board, and tool integrations. Agents join as named seats, keep their own identity and runtime, and can be connected via a daemon, manual CLI, or MCP package. The system includes multi‑LLM routing, GitHub‑synced tasks, scoped grants for external services, and a marketplace of pre‑built agents. It runs on Docker‑Compose locally or via Helm on Kubernetes, is Apache‑2.0 licensed, and provides extensive docs and community support.

  • Project
    TencentCloud/TencentDB-Agent-Memory

    TencentDB Agent Memory is an open‑source memory hub for LLM agents. It captures conversations, extracts reusable skills, builds a Wiki from docs, and indexes code into a CodeGraph. A central UI (Memory Hub) lets teams manage assets, set fine‑grained ACLs, and bind specific memory pieces to individual agents. Agents talk to a uniform proxy, so any framework (Claude Code, DeepSeek Harness, OpenClaw, etc.) can use the same memory without code changes. Benchmarks show a 59 % boost in persona retention. The repo provides one‑click Docker deployment, migration tools, and a roadmap for zero‑config cold starts and more integrations.

  • Project
    lobehub/lobehub

    LobeHub is an open‑source platform for orchestrating AI agents as a 24/7 “team”. It lets you create agents via a builder, give them access to any LLM and thousands of plugins, schedule and monitor their work, and store editable personal memory. Deployable via Vercel, Zeabur, Alibaba Cloud, or a Docker image, it requires only an OpenAI API key. The project also provides UI libraries for building AIGC web apps.

  • Project
    Fergana-Labs/stash

    Stash is an open‑source platform that captures AI‑agent runs, curates the resulting traces into durable knowledge, and lets you package that knowledge as reusable “skills”. It offers a Python SDK, REST API, CLI, and an optional reward‑model loop that turns human‑annotated preferences into prompt‑optimisation signals, enabling continuous improvement of agents without retraining the underlying model.

  • Project
    HKUDS/AgentSpace

    An agent-native collaborative workspace that transforms AI agents into governed digital employees for human-agent teams to collaborate in a shared environment.