EntwineLLM
About
EntwineLLM exists for developers whose code cannot leave the machine — regulated environments, air-gapped networks, or simply preference for local inference. The extension sends selected code to an Ollama or LMStudio endpoint (local or Docker-hosted, with bearer-token auth added in v1.13 for proxied setups) and returns refactoring suggestions aligned with Clean Code principles, unit tests covering execution paths, Markdown documentation exportable to HTML, or a code review from a senior-developer perspective. Follow-up prompts allow iterative refinement, and per-command model and timeout settings let users route cheap tasks to smaller local models. It supports C#, Python, and Java, and is a single-maintainer project with modest but continuing releases.