holaos is a local-first, all-in-one AI agent workspace that lets users run any agent side-by-side with shared memory, tools, and files. Context, preferences, and project history are stored locally as plain files and shared across all agents and sessions.
Project overview
By keeping context, preferences, and project history in local plain files shared across every agent and session, holaos eliminates the context loss and setup rebuilding inherent in fragmented agent workflows.
Project type
MCP · AI Agent · AI Coding
Use cases
Coding & Development
Deployment
Refer to project documentation
License
License pending
Best for
Developers and enterprise teams seeking a local-first, all-in-one workspace to run multiple AI agents side-by-side with shared memory and standard-format deliverable generation.
Users who need context, preferences, and project history stored locally as plain files and shared across all agents and sessions.
Key capabilities
Context, preferences, and project history are stored locally as plain files and shared across all agents and sessions.
Users can install apps from the in-workspace marketplace; these apps open as real, interactive surfaces right beside the agent.
Agents can drive browsers, generate media, and produce real deliverables saved as xlsx, pptx, and docx files.
Limitations and risks
holaos uses a Modified Apache 2.0 license with commercial-distribution and branding conditions. Users should review these conditions before adoption.
Beyond built-in frontier models, users can optionally connect external LLM API providers via BYOK; paid services may be involved depending on the provider chosen.
Getting started
A desktop app download is available with no setup required; frontier models are built in and free to start. A one-line install script is also available for bootstrap. No coding is required to begin.