Claude Code cut the status-line tax by ~99%
SummaryBackground subagent status used to re-read the whole chat every 30 seconds. Now Haiku 5.5 writes it from short notes — Anthropic says each update costs about 99% less, and sessions with background subagents used ~5% less of the usage limit (~6% more prompts). CLI v2.1.316+.
· MODEL Editors · 1 min read
Some links may be affiliate links. Disclosure
Anthropic just fixed a quiet Claude Code bill: the background subagent status line.
What was wrong. While a background subagent ran, the CLI status line refreshed about every 30 seconds. Each refresh re-read the whole conversation just to print something tiny — like "reading auth tests." Same job, full-context cost. Call it a subscription tax on progress theater.
What changed. The status author is now Haiku 5.5, writing from short notes instead of the full chat. Anthropic / ClaudeDevs (opens in a new tab) say each status update costs about 99% less. Sessions that use background subagents used roughly 5% less of the usage limit — about 6% more prompts in the same budget. MODEL did not independently measure the 99% / 5% / 6% figures; those are Anthropic's.
What did not change. The subagent work model is the same. Only who writes the status line changed. This is CLI-only (desktop does not write this status line). Live for most Pro / Max / Ultra and all Team / Enterprise. Update to v2.1.316+.
Why it matters. Tiny UI strings that re-ingest the transcript every half-minute add up on paid plans. Cheaper status means more of your limit goes to actual agent work — not narration.
The catch. Still Anthropic's numbers until outsiders replicate them. Desktop users waiting for a matching status path get nothing new here. And a cheaper "still reading" line does not make the subagent faster — it just stops charging full-chat rates to say so.
Sources: ClaudeDevs Nov 4 (opens in a new tab); ClaudeDevs follow-up (opens in a new tab); Pasquale Pillitteri write-up (opens in a new tab).
MODEL Editors
Sources
- Claude Code status line cost cut (Nov 4, 2026)
ClaudeDevs (X)
- Follow-up on Haiku 5.5 status author (Nov 4, 2026)
ClaudeDevs (X)
- Claude Code Subagent Status Line Usage Limits
Pasquale Pillitteri
Keep reading
AI News
Claude can now read your bank accounts (read-only)
On Nov 5, 2026 Anthropic launched Personal Finance in Claude with Plaid: connect US bank and card accounts, ask in natural language, read-only access. Beta this week for Pro, Max, and Ultra in the US — gradual rollout, paid plan does not mean instant access.
· MODEL Editors
AI News
OpenAI says it will pay its own data-center power bill
On Nov 5, 2026 OpenAI published a National Community Compact for campuses it owns and operates, plus a Compute Campus Policy Blueprint. CNBC, citing Ballotpedia: of 41 local data-center measures across 10 states on Nov 3, voters backed restrictions or rejected pro-development measures in all but one.
· MODEL Editors
AI News
Apple put AI at the center of everything. Then it put engineers on a $200/day AI allowance.
Bloomberg via MacRumors (Nov 4): some Apple engineers cut from $300/day AI spend to $200 in October; further cuts toward $100 for some eng and $25 for some non-eng discussed. Claude, ChatGPT, Gemini allowed. Employees can seek executive exemptions. $AAPL. Not investment advice.
· MODEL Editors
FREE · EVERY WEEK
The top AI tools this week, tested.
Every week MODEL's editors pick the AI tools worth your time, try them hands-on, and tell you what's good, what's not, and who it's for. Free, in your inbox.
Free. Unsubscribe anytime. Our picks are editorial and can't be bought.