context management

Appending every turn is the default in tutorial code and it fails twice: on the context window and on cost. The signal is the layered assembly, the constraints you refuse to let the summarizer touch, and what you do the moment the user changes subject.

An emerging alternative to stuffing the window or building an index: keep the context outside the prompt and let the model write code to inspect it, recursing into the pieces. Knowing when it does not pay is most of the score.

Conversations and agent tasks grow past the context window, and the naive 'stuff the whole history' approach breaks on cost, latency, and lost-in-the-middle. The signal is a tiered memory design: recent buffer, summarized mid-term, retrieved long-term.

Agents need memory beyond the context window, and 'just store the history' is the wrong answer. The signal is telling the memory types apart and mapping each to a storage and retrieval strategy. Here is the answer.

The agent loop assumes an observation is a short string, and images break that assumption in both directions. The signal is a context policy that keeps pixels out of the transcript and a verification step for the files the agent produces.