NaSh: Guardrails for an LLM-Powered Natural Language Shell
This addresses usability and safety issues for users of LLM-based command-line interfaces, but it is incremental as it builds on existing shell and LLM concepts.
The paper tackles the problem of unintended or unexplainable outputs from LLM-powered natural language shells by designing NaSh, a new shell with guardrails to help users recover from errors, though no concrete performance numbers are provided.
We explore how a shell that uses an LLM to accept natural language input might be designed differently from the shells of today. As LLMs may produce unintended or unexplainable outputs, we argue that a natural language shell should provide guardrails that empower users to recover from such errors. We concretize some ideas for doing so by designing a new shell called NaSh, identify remaining open problems in this space, and discuss research directions to address them.