I kept running into the same problem: I'd solve an issue or make a technical decision, and a few weeks later I had to search through old AI conversations, Git history, and notes to recover the context behind it.
So I built Kote. It passively captures development context from AI sessions and Git activity and makes it available while working on the code.
The main feature is a VS Code CodeLens integration that surfaces relevant historical context and technical decisions when you open a file, allowing you to retrieve the context associated with that code directly in the editor.
The indexed context is also searchable through the web app, CLI, and MCP. Kote can be self-hosted and supports using your own LLM provider.
https://github.com/pedroaugusto04/Kote How do you currently preserve or retrieve the context behind technical decisions in your projects?