Panes

Starts every conversation oriented: one context block with the repo, branch, layout, stack and its build/test/lint/run commands, README summary and recent commits.

Every prompt you write, kept across sessions and projects: /history searches them and puts one back in the prompt box.

Text expansion in the prompt box: type ;plan and a space, get the full instruction.

Blocks catastrophic shell commands (rm -rf ~, curl | sh, force-push to main, DROP DATABASE, format C:) and asks before risky ones (git reset --hard, rm -rf, terraform apply), in Bash and PowerShell.

Notices the commands you keep approving by hand and offers a one-click standing allow rule (never for risky or interpreter-wide commands).

Commit messages in your repository's own style (learned from its last 30 subjects) and pull request descriptions from the branch, drafted by haiku; nothing is committed or pushed without a press.

Opt-in: questions about the codebase (where is, what does, how does, which file, find) go to a read-only haiku scout subagent first; its answer comes with Ask Claude anyway and Good enough, and the tokens saved.

During long turns, haiku writes one line above the prompt every ~30 s on what Claude is doing and what comes next, from a compact log of its tool calls; a one-line recap toast after turns over two minutes.

A side conversation with haiku or sonnet in a pane, kept out of Claude's context: attach Claude's last answer, the diff, the plan or a file on demand, then send the answer or a summary to the prompt box.

Opt-in: after a turn that adds or changes functions, sonnet (or haiku) drafts focused unit tests in the background, in the repo's framework and style; Review, Add & run (only that file), or Dismiss; never writes without a press; daily budget cap.

Prompt-cache hit rate, a live warm/cold countdown, cache-bust alerts with likely causes, savings, and optional keep-warm pings.

Context-window gauge: a fill bar in the readout, toasts as it fills, a /ctx breakdown pane with Compact now, and optional auto-compaction between turns.

Spend tracking across sessions: session and today's cost in the readout, daily and weekly budgets with toasts and an optional hard stop, and a /spend pane.

Subscription rate limits at a glance: each window and its reset in the readout, toasts at 75% and 90% and when your pace beats the reset, and a /limits pane.

Every subagent, live: a /subagents pane with status, model, elapsed time, steps, tokens, last tool and answer, plus a band while any run.

Tool latency: times every tool call, shows a live timer for slow ones, and a /stopwatch pane with per-tool averages, p95, max and the slowest calls.

A receipt for every turn: tools, tokens, cache hits, estimated cost and lines changed, on the turn's closing line (or in the transcript), with a /receipts pane.

A /dashboard pane charting this session's turns (prompt tokens by cache kind, hit rate, cost) and your last 14 days of spend.

Picks the model for each subagent by its type (Explore on Haiku by default) and can cap how many run at once.

Per-prompt effort autopilot: classifies each prompt (Jev, Claude Haiku, or heuristics) and sets that turn's effort, and optionally its model, only when the switch beats the prompt-cache rewrite it causes.

Asks before /model, /output-style, /fast, /effort, MCP or plugin changes and /config rows that would throw away a warm prompt cache, with the cost of rewriting it; names the rebuilds it could not prevent.

Forecasts how many turns remain until auto-compaction from smoothed per-turn context growth (tool output vs conversation), and at natural breaks offers Compact now with a focus Haiku drafts.

See the system prompt section by section with sizes, and switch sections off for a project.

See the reminders and notices Claude Code injects for the model, by type and size, and mute the ones you do not want.