{"version":1,"type":"story","url":"https://digestai.news/story/claude-code-burns-about-1520-of-tokens-on-3word-status-line","json":"https://digestai.news/story/claude-code-burns-about-1520-of-tokens-on-3word-status-line.json","markdown":"https://digestai.news/story/claude-code-burns-about-1520-of-tokens-on-3word-status-line.md","slug":"claude-code-burns-about-1520-of-tokens-on-3word-status-line","headline":"Claude Code burns about 15–20% of tokens on 3‑word status line","summary":"Flamerix AI reports that Claude Code sends a background‑agent summary request roughly every 30 seconds. The request copies the entire transcript to the agent’s model (usually Opus) and returns a three‑to‑five‑word status line. Because the request is billed like any other Claude Code call, it can consume around 15–20% of a user’s token budget, reducing rate‑limit headroom.\n\nIndependent measurements collected by 86879 show an overhead of 18–26 %. The author’s own telemetry on version 2.1.301 recorded a weekly spend share that grew from 3.5–7 % in July to 7.3 % then 11.8 % the following week, reaching about 16 % by mid‑October. The cost is mostly cached reads, but the uncached 2–3 % of calls carry a third of the expense.\n\nA workaround is provided via a Bun preload script (no‑agent‑summary.js) that intercepts the timer callback and disables the summary request. Installing the script through the BUNOPTIONS environment variable (or a wrapper) stops the extra calls without altering the binary, restoring normal token usage.","keyPoints":["Claude Code sends a summary request every ~30 seconds, using the full transcript and costing ~15–20% of token budget.","Independent measurements report 18–26% overhead; weekly spend share rose to ~16% by mid‑October.","A Bun preload script (no‑agent‑summary.js) can suppress the timer, eliminating the extra token usage."],"whyItMatters":"The hidden token overhead cuts into rate‑limit headroom and raises costs for Claude Code users, especially in long‑running or background tasks, so removing it improves efficiency and budget predictability.","category":{"slug":"agents","name":"Agents & Tools","url":"https://digestai.news/category/agents"},"entities":{"companies":["Anthropic","Flamerix AI"],"models":["Claude Code"],"people":[]},"firstPublishedAt":"2026-10-29T03:33:42Z","updatedAt":"2026-10-29T03:33:42Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"gist.github.com","title":"Claude Code burns ~15–20% of your tokens on a 3-word status line","url":"https://gist.github.com/flamerix-ai/c3905705cc4205e5de3011fa2e74efce","publishedAt":"2026-10-29T03:33:42Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[{"site":"Hacker News","url":"https://news.ycombinator.com/item?id=50284255","points":88}],"thread":{"title":"Claude Code Token Efficiency Field Test","url":"https://digestai.news/thread/claude-code-vs-cursor-field-test-shows-token-usage-differs","storyCount":2},"cite":{"text":"Digest AI, \"Claude Code burns about 15–20% of tokens on 3‑word status line\", 29 October 2026, https://digestai.news/story/claude-code-burns-about-1520-of-tokens-on-3word-status-line","publisher":"Digest AI","title":"Claude Code burns about 15–20% of tokens on 3‑word status line","datePublished":"2026-10-29T03:33:42Z","url":"https://digestai.news/story/claude-code-burns-about-1520-of-tokens-on-3word-status-line"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}