Most AI coding workflows treat context as something that happens accidentally. You open a few files. Paste a stack trace. Ask the model to inspect a directory. Then another. Then the chat grows heavy, the model starts missing earlier instructions, and everyone pretends the problem is "the model got weird." In ai-assistant-dot-files (I probably need to rename this now as its grown into something else) I wanted to treat the context window as a budget. Not a vibe. Not a giant bucket. A b...