Lawrence Jengar
Apr 16, 2026 00:53
Anthropic releases /utilization command and context administration information for Claude Code, serving to builders navigate 1 million token classes extra successfully.
Anthropic dropped a brand new /utilization command for Claude Code on April 15, alongside detailed steerage on managing classes inside the AI coding assistant’s a million token context window. The replace addresses a rising ache level builders have flagged: determining when to proceed, compact, or nuke a session completely.
The corporate’s technical employees member Thariq Shihipar laid out the logic in a weblog put up that reads extra like tribal data than typical documentation. The core downside? Context rot—the place mannequin efficiency degrades as extra tokens pile up and a focus will get unfold skinny throughout irrelevant cruft.
5 Methods to Deal with Context Bloat
Claude Code now affords builders a choice tree at each flip:
Proceed works when every thing within the window nonetheless issues. /rewind (or double-tap Esc) enables you to bounce again to any earlier message and reprompt, dropping every thing after that time. Anthropic suggests this beats the intuition to sort “that did not work, strive X as a substitute”—as a substitute, rewind to only after helpful file reads and provides cleaner directions.
/compact asks Claude to summarize the session and substitute historical past with that abstract. It is lossy however automated. You possibly can steer it: “/compact concentrate on the auth refactor, drop the take a look at debugging.” The catch? Autocompact fires when context is almost full, exactly when the mannequin performs worst because of context rot.
/clear places you in management. You write what issues and begin contemporary. Extra work, cleaner slate.
Subagents deal with the messiest state of affairs—when the subsequent chunk of labor will generate tons of intermediate output you will not want once more. Spawning a subagent provides it a contemporary context window; solely the ultimate report comes again to the dad or mum session.
The Psychological Mannequin
Anthropic’s inside take a look at: “Will I would like this instrument output once more, or simply the conclusion?” If simply the conclusion, spin up a subagent for codebase searches, verification runs, or doc writing.
The steerage suggests beginning contemporary classes for genuinely new duties, regardless that 1M context technically permits marathon classes. For associated follow-up work—like documenting a characteristic you simply constructed—staying in session avoids costly file rereads.
Dangerous autocompacts usually occur when Claude cannot predict the place your work is heading. An extended debugging session will get summarized, then you definitely ask about “that different warning in bar.ts” that bought dropped from the abstract. With the bigger context window, Anthropic recommends proactive /compact with express path earlier than the automated set off fires.
The replace displays Anthropic’s push to make Claude Code extra sensible for prolonged improvement classes. Whether or not this strikes the needle on enterprise adoption stays to be seen, however builders burning via context on advanced codebases lastly have documented methods past trial and error.
Picture supply: Shutterstock
