| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| caprock_0.53.0_windows_amd64.zip | 2026-09-06 | 8.9 MB | |
| caprock_0.53.0_windows_arm64.zip | 2026-09-06 | 8.1 MB | |
| checksums.txt | 2026-09-06 | 600 Bytes | |
| caprock_0.53.0_darwin_amd64.tar.gz | 2026-09-06 | 8.8 MB | |
| caprock_0.53.0_darwin_arm64.tar.gz | 2026-09-06 | 8.3 MB | |
| caprock_0.53.0_linux_amd64.tar.gz | 2026-09-06 | 8.7 MB | |
| caprock_0.53.0_linux_arm64.tar.gz | 2026-09-06 | 8.1 MB | |
| README.md | 2026-09-06 | 1.6 kB | |
| v0.53.0 source code.tar.gz | 2026-09-06 | 6.3 MB | |
| v0.53.0 source code.zip | 2026-09-06 | 6.6 MB | |
| Totals: 10 Items | 63.7 MB | 0 | |
Added
-
The status line carries the mark — the amber ⛰ from the favicon, the glyph alone, so the line is identifiably ours instead of an anonymous row of figures. Two columns, and it is never dropped however narrow the terminal gets. This is on every status line, in both modes, and needs nothing turned on: if Caprock's status line is already set up, upgrading is enough.
-
The session's own numbers in the Claude Code status line —
N turns · N steps · cache −N% · in N · out N, from counters only the daemon has, added to the line Caprock already prints.
Nothing to turn on, and nothing to re-register: if Caprock's status line is
already set up, upgrading is enough, and your settings are not touched. New
installs get it from caprock up as before. caprock statusline --plain
opts out.
The daemon read that fetches the counters is bounded at 40ms — measured against a real endpoint that answers in 0.6ms — and any failure falls through to the line you get today. A stopped daemon costs nothing measurable; a hung one costs ~55ms per line.
It is fitted to the terminal — the counters go one at a time, and the plan windows are never dropped, being the number that decides whether work can continue at all.
Not shown, deliberately: time to first token, tokens per second, and the model-versus-tool time split. Nothing carries them — not the hooks, not the transcript — and the obvious derivation from hook timestamps counts your own thinking time as inference. They stay off the line until something real backs them.