Originally created by @ghost on GitHub (Mar 30, 2024).
Describe the solution you'd like
Establish modular design, requirement, process, security, best practice, etc. With all of the innovation surrounding LLM technology, software like code execution & extended RAG can be extremely beneficial but not without capacity and security considerations. Candidates like open-interpreter and self-operating-computer are exciting!
My hope is that we can keep the core of Open-WebUI secure, lightweight, and performant. One solution may be modularity. I'd love to see capability like code execution available as an opt-in module with logical and secure separation. If we can establish guidelines for Open-WebUI Modules, this can simplify the development of future modules.
Originally created by @ghost on GitHub (Mar 30, 2024).
**Describe the solution you'd like**
Establish modular design, requirement, process, security, best practice, etc. With all of the innovation surrounding LLM technology, software like [code execution](https://github.com/open-webui/open-webui/issues/1321) & [extended RAG](https://github.com/open-webui/pipelines/issues/15) can be extremely beneficial but not without capacity and security considerations. Candidates like [open-interpreter](https://github.com/OpenInterpreter/open-interpreter) and [self-operating-computer](https://github.com/OthersideAI/self-operating-computer) are exciting!
My hope is that we can keep the core of Open-WebUI secure, lightweight, and performant. One solution may be modularity. I'd love to see capability like _code execution_ available as an opt-in module with logical and secure separation. If we can establish guidelines for **Open-WebUI Modules**, this can simplify the development of future modules.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @ghost on GitHub (Mar 30, 2024).
Describe the solution you'd like
Establish modular design, requirement, process, security, best practice, etc. With all of the innovation surrounding LLM technology, software like code execution & extended RAG can be extremely beneficial but not without capacity and security considerations. Candidates like open-interpreter and self-operating-computer are exciting!
My hope is that we can keep the core of Open-WebUI secure, lightweight, and performant. One solution may be modularity. I'd love to see capability like code execution available as an opt-in module with logical and secure separation. If we can establish guidelines for Open-WebUI Modules, this can simplify the development of future modules.
@tjbck commented on GitHub (Jun 1, 2024):
Related: https://github.com/open-webui/pipelines