Nedster - Private local AI coding agent for Ollama

by
Nedster turns Ollama into a practical terminal coding agent. It detects your hardware, recommends local models, retrieves project context, edits files, runs tools and tests, and verifies behavior before declaring success. Routine inference and project data stay on your machine by default. One installer creates a global command for Windows, macOS, and Linux.

Add a comment

Replies

Best
Maker
📌
I built Nedster because local models are private and increasingly capable, but a model alone is not a dependable coding agent. The missing piece is a disciplined harness: relevant project context, constrained tools, structured plans, feedback from real failures, and verification after every change. Nedster packages that into one terminal command and chooses a sensible Ollama model fleet for the machine it runs on. This is Alpha 2, so I’d love candid feedback on setup, local-model reliability, and the balance between a quiet interface and useful execution detail.