A self-hosted, provider-agnostic web interface for chatting with local Ollama models and OpenAI-compatible APIs, with RAG, plugins and multi-user access control.
- Stars
- 39,802
- Forks
- 6,242
- Open issues
- 8
- Last commit
- 20 hours ago
What is it?
MindsHub is an agent workspace for research, analysis, content creation and software development. It lets users pick among frontier and open models from multiple providers rather than being tied to a single one, and it supports interchangeable open-source agent harnesses such as Anton and Hermes.
The repository is a superproject that bundles the desktop and web app, the agent backend and the data engine, so the full stack can be built and run from source on macOS, Windows or Linux. The project also covers connected data sources through a credential vault, scheduled tasks, cross-session memory, a reusable skill library, and the ability to publish generated documents, dashboards and apps to a live URL.
Deployment options include the hosted web app, a local desktop install, or self-managed setups in a VPC, on-prem or air-gapped environments.
Key features
- 01Choose among frontier and open models from multiple providers through a Model Router.
- 02Swap between open-source agent harnesses such as Anton and Hermes.
- 03Connect data sources like Postgres, BigQuery, Gmail and Notion through a scoped credential vault.
- 04Turn agent output into documents, dashboards, apps and code, published to a live URL.
- 05Run tasks on a schedule with cross-session memory and a reusable skill library.
Use cases
- Automating recurring reporting and monitoring tasks that read and write documents.
- Building internal dashboards or small apps for a team without a full engineering setup.
- Running an agent workspace on-premise or in a private VPC to keep data under your control.
- Comparing and switching between different LLM providers within one interface.
Quick start
- 1
Clone the repository with submodules
git clone --recurse-submodules https://github.com/mindsdb/minds.git - 2
Enter the project directory
cd minds - 3
Install dependencies
make setup - 4
Run the desktop app with hot reload
make dev
Things to know
- Linux has no prebuilt app and requires building from source.
- Free tier is limited; frontier models and private artifacts need a Pro plan.
- Running the full stack from source requires several modules and local tooling.
Topics
FAQ
- What is mindshub?
- An agent workspace for knowledge work and software development, offering a choice of models and open-source agent harnesses, runnable locally or hosted. MindsHub is an agent workspace for research, analysis, content creation and software development. It lets users pick among frontier and open models from multiple providers rather than being tied to a single one, and it supports interchangeable open-source agent harnesses such as Anton and Hermes.
- Is mindshub free to use? What license does it have?
- mindshub is open source under the MIT license. Check the license terms before using it in commercial products.
- Is mindshub still maintained?
- Actively maintained. The last commit to mindshub was on Oct 19, 2026.
- What can I use mindshub for?
- Automating recurring reporting and monitoring tasks that read and write documents. Building internal dashboards or small apps for a team without a full engineering setup. Running an agent workspace on-premise or in a private VPC to keep data under your control. Comparing and switching between different LLM providers within one interface.
- How do I install mindshub?
- Clone the repository with submodules: git clone --recurse-submodules https://github.com/mindsdb/minds.git Enter the project directory: cd minds Install dependencies: make setup Run the desktop app with hot reload: make dev
- Where can I get mindshub?
- The source code and installation instructions are on GitHub: https://github.com/mindsdb/mindshub
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
LobeHub is a platform for organizing AI agents into always-on teams that can be hired, scheduled, and reviewed from one place.
Open-source personal AI assistant and Agent Harness that plans tasks, runs tools and skills, and keeps memory across web and IM channels.
Self-hosted AI chat and knowledge platform that indexes data from 50+ connected apps and works with most LLM providers.
A minimal Python library for building AI agents that write and run their actions as Python code.
A TypeScript framework for building AI agents and LLM-powered applications, with workflows, tools, memory, MCP servers, and evals.