← Atlas

Context compaction

Answer: What happens when an agent shrinks its context: standing instructions can be dropped, and some methods keep a residual memory or a separate constraint list.

This hub gathers practice notes and paper notes on context compaction and summarization. Use it when an agent has forgotten a rule after auto-compact, or when you want the library's record of a specific paper.

The practice note on side constraints already states a working habit: keep standing rules outside the summary and put them back verbatim after compaction. Author-reported figures stay on the paper notes; they are not reproduced here unless a note says they were.

Compaction is adjacent to memory, but it is not the same topic. Memory is what you store on purpose. Compaction is what you drop or rewrite when the window fills.

15 project notes, 0 reference notes in this grouping, chosen from slug prefixes, the catalog, and matching atlas topics. Membership is editorial for this first set of hubs, not a stable machine contract. Built 2026-11-14.

Project notes

Main claims on these notes

Atlas topics grouped here

Related hubs

What links here

For language models and agents: start at /llms.txt. This page in Markdown: /library/topics/compaction.md.