Show HN: A laboratory to study how LLMs behave when offered a set of fake tools

Category: devtools

Tags: llm-testing, tool-calling, experimentation

Score: 5.3/10 (Innovation: 5, Technical: 6, Documentation: 6, Utility: 4)

Llaboratory is a self-hostable tool for creating and running experiments where LLMs interact with fake, configurable tools. It enables studying model behavior and tool-calling patterns in a controlled environment, which is interesting for researchers and developers working with LLM agents.

Target audience: AI researchers, ML engineers, backend devs

Repository: https://github.com/ampyard/Llaboratory · TypeScript · MIT

View on Hacker News