Back to directory

holaOS

An open-source, local-first agentic workspace that enterprises can shape to their own tools, connecting existing chat apps and 100-plus integrations to any coding agent or model, with shared memory.

holaOS

holaOS is an open-source, local-first agentic workspace that a company assembles around how its work actually runs, rather than adopting a single fixed agent product. It connects the chat tools and systems a team already uses, more than 100 integrations plus MCP, and works with any coding agent, such as Claude Code or Codex, and any model, including a user's own API key, with all of it running on the team's own machines rather than a vendor's cloud.

Its app feature installs apps from an in-workspace marketplace as real, interactive interfaces, such as Notion or a browser, that sit next to the agent rather than as a chat transcript, so a person can watch the agent drive the app, take over at any point, and have their own manual actions become context the agent already has. A separate messaging feature connects a team's existing chat tools, such as Slack, Feishu, DingTalk and WeChat, so the agent can read decisions and requirements directly from the conversations where they were made, under the user's control over what it can see.

Setup is described as taking clicks rather than months.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.