A new open-source tool called Bash4LLM provides a lightweight, dependency-free way to interact with large language models directly from the Unix terminal. Created by developer Kamaludu, the wrapper consists of a single Bash script that leverages only Bash, curl, and jq — eliminating the need for Python, Node, or other runtimes.
The tool supports sending prompts, conducting short chat sessions, processing files line by line, streaming output, and saving session metadata in JSON format. Its design prioritizes safety and predictability: it avoids using the system /tmp directory and never calls eval. Groq is supported out of the box, while other providers can be integrated via dedicated Bash scripts placed in the extras/providers/ folder.
According to the developer, the project was born from a desire for simplicity. "I created it because I wanted something simple that worked without installing Python, Node, or any other runtime," Kamaludu states. The tool's minimal footprint makes it particularly appealing for resource-constrained environments or users who prefer staying within the shell.
For now, Bash4LLM remains a niche offering compared to more feature-rich CLI clients like Ollama or llm-cli. Its reliance on jq for JSON processing and curl for network requests means basic Unix skill requirements remain modest, though users must have these utilities installed.
Community reaction on Hacker News has been measured, with the submission receiving 7 points and 6 comments so far. The tool's extensibility through provider scripts could broaden its appeal over time.