What Claude Code's 30-second agent summary call costs, read from the binary
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
A Hacker News story reports that Claude Code sends each background agent's transcript to the model roughly every 30 seconds to produce a 3-to-5-word status line, and that this can account for a sizable share of spend.