The Four Habits Eating Your Claude Limit, and the Fix for Each
Claude's /usage screen now warns about max effort, workflows, always-on context and large tool results. What each one is, the exact fix for each, and what the numbers can and can't tell you
On October 14, Claude Code 2.1.299 added warnings in /usage for four habits that burn through a plan: max effort, workflows, always-on context and large tool results. Run /usage, read which one shows up, and fix that one first.
Here's what each habit is, the fix for it, and the catch about the numbers.
Why usage climbs
- Claude Code sends your whole conversation with every request, and each time Claude uses tools it sends another request carrying that batch of tool results.
- Every subagent, and every agent a workflow spawns, sends its own requests on top of the main conversation's.
Fix 1: max effort
- Lower the effort for simple work: run /effort, or change it in /model.
- Give a helper its own level: add effort: low to its file in .claude/agents/.
- Heads up: the CLAUDE_CODE_EFFORT_LEVEL environment variable overrides both, so unset it if the level won't change.
Fix 2: workflows
- Set CLAUDE_CODE_WORKFLOW_SUBAGENT_MODEL to run every workflow agent on one cheaper model while your other subagents keep theirs (needs v2.1.296 or later).
- For ordinary helpers, put model: haiku in the helper's file.
Free, for your email
Get the full setup steps + the copy-paste setup prompt
The rest of this guide (4 more sections, 1 copy-paste prompt) opens right here. The setup prompt, one paste into Claude that sets it all up, comes by email only. Use your real address.
Unlocks: Fix 3: always-on context · Fix 4: large tool results · The catch · Sources
50,000+ follow where these guides come from. No spam, reply “unsubscribe” any time.