Bands

Reads Claude's last response aloud: natural edge-tts voices (free, no API key) with an offline fallback to the OS voice, plus pause and stop buttons.

Progress bar above the prompt for /savvy-flow or any delegated plan, plus an agents panel (/agents-info): running, finished and planned subagents with model, task progress, context, cost and time. Based on johnnyvizz/claude-kit savvy-progress.

Suggests one of your skills from what you are working on: your message, the last few turns, the files Claude reads and writes, project hints and your past clicks. Local BM25 over each SKILL.md; no model calls, no tokens until you click.

Offers /handoff-journal or /handoff-local above the prompt when the context passes 60% and 85%, or the session runs past 3 hours. Reads only the engine's own usage figures: no tokens until you click.

Whenever nothing is running (session start, after a reply, after a stopped reply), shows a checkbox panel of next-step options above the prompt. Tick one or more and Send selected or Edit selected, or send one with a click.

A panda colleague in a side pane who works along with each task and keeps a short management summary of the session.

Switchboard: useful extras for Claude Code, each one a switch. A usage meter above the prompt, an output viewer for long results and images, secret hiding, a stop for repeated failures, a check before risky commands, a work log, and coordination between parallel chats. /switchboard turns each on or off.

A permanent pane: the ticket and summary of what Claude works on, its progress, and its todo list

Anexa a skill do seu design system ao primeiro pedido de peça visual da sessão. Configure o nome da skill para ligar.

Mantém um servidor de prévia no ar para o HTML editado e mostra a URL acima do prompt.

A permanent, slim context-usage bar above the prompt with a chomping Pac-Man; green under 20%, amber 20-30%, red past 30%, munching faster each zone, with a fun line per tier

A usage meter for Claude Code: a plain "5h 62%" button under the prompt that opens a compact sheet with your session, weekly and per-model weekly limits, and how long you have left at your current pace.

A row of pills above the prompt: rate limits with pace, context, tokens and cost.

Shows what actually ran next to what Claude said ran: checks, stale passes, and unbacked claims.

Warp-style Rich Input for Claude Code: the prompt box as a multi-line editor you switch on and off (Enter for a new line, Ctrl+Enter to send), an editor pane for mouse-capable terminals, and @ completion that browses folder by folder

Shows every agent of the session, the model it ran on and its tokens; /agents opens the pane

Triages each message with a small model before the main model runs, and asks (native modal) whether to switch model/effort when another one fits better and the switch pays off.

Bitbucket review queue and your open PRs in a pane; the current branch's PR, CI and Jira issue in a bar above the prompt

三欄待辦看板（代辦、進行中、完成），資料存在 Notion 資料庫或只存本機；整合 Herdr 分頁與 cc-mod-park

Notifies you X minutes before your prompt cache expires so you can send a turn before paying the full re-cache cost.

Turns a session into an orchestrator that delegates work to haiku and sonnet subagents, keeps its own context small, sizes each task to about 150K tokens of subagent context, and lists the running and past agents in an expandable band (/orchestrator on|off|status)

A tiny ASCII pet above your prompt (=^.^=) that cheers when tests pass, worries when commands fail, naps when you are idle, and levels up as you work.

Kollisionsschutz für parallele Sitzungen, Kontext-Anzeige mit „Stand sichern“ und Brain-Band — 0 Tokens, nur lokal.

Configurable band above the Claude Code prompt: context, cache, quota, model, git and more, with a details pane and an in-app config editor.