client-clock
claudeLogs your time, Claude's time and your 5-hour/weekly usage per client, so client work is easy to bill and budget
Claude Code and Codex plugins their authors filed under "productivity". All categories.
Logs your time, Claude's time and your 5-hour/weekly usage per client, so client work is easy to bill and budget
A tape of every edit Claude makes: replay each change typing itself in, step through it, and rewind any file to any step.
Opt-in ads for Claude Code: a banner above the prompt and a Theater pane while Claude works. Ads never enter the model's context. Unofficial, not affiliated with Anthropic.
Keeps the prompt cache warm while you step away with a short poke before it expires, and wakes Claude when the usage limit resets. A coffee cup above the prompt shows how warm the cache is.
A list of what to look at in your browser after Claude changes something you can see, each with a link, a tick and a way to say it is not right.
A pane that shows where the current branch is in the CI/CD pipeline, from local dev to prod smoke tests
A box above the prompt that lists the actions and questions the agent left for you
/rewrite: turns a rough draft prompt into a thorough one, asking clarifying questions first, without adding anything to the conversation
Cost-free mode: helper agents run on Haiku, and big files and logs are summarised by a free Gemini model instead of filling Claude's context.
When a reply contains a command for the user to run (a code block starting with "! "), copies it to the clipboard and places it in the prompt input if empty. Pressing Enter on that exact text runs it and feeds the output back to the model.
Zijpaneel en statusregel met de voortgang van de Jira-Epic waar het ticket van je sessie onder valt (ring, balk, tickets, ≈ sessies te gaan), met /epic-commando. Leest Jira via de Atlassian-connector.
A progress band with a little coral friend for Claude Code: task and subtask lines, a gradient progress bar, subagent helpers and moods that follow what Claude is doing.
A task band above the prompt that opens by itself: Claude plans multi-step work, ticks each task as it goes, and the band closes when everything is done.
A variant of next-steps: tick several suggested next prompts, in the order you choose, and run them as a sequence, one step per turn.
The control panel for Claude Code: it sets up each project with the right skills and tools, shows what is in use, and vets new tools before you add them.
A one-line usage bar above the prompt: 5-hour, weekly and per-model plan limits with reset countdowns, plus the session's tokens and cost.
Rate limits, context, cost and git branch in a band above the prompt
Answers Claude's questions with TypeSafe Jev so autonomous loops don't stall; escalates big calls to Telegram and logs every decision
A card above the prompt: the model, the context window in tapered colour bands, prompt cache warmth, and the git repo and branch.
Bento grid in the side pane: clock, todo, context forecast, git, Hacker News.
The multi-AI toolbar for Claude Code: usage gauges, one-click model buttons, Sub5 parallel sub agents in git worktrees, read-only delegate buttons for Codex, Cursor agent and agy, a plain-language Recap, a tabbed settings pane, opt-in translation and web search by a delegate, and deploy watch, Claude-Codex handoff and attribution/polling guards. In 41 languages.
Live emoji forecast of your context window above the prompt, with a one-click /handoff to a fresh context.
Progress bar and agents panel with pixel crabs, on the desktop and in any truecolor terminal (fork of johnnyvizz/claude-kit, MIT)
Keeps the savvy-progress bar on for every task
Ships one output style combining Simplified Technical English (short, direct sentences) with an ADHD-friendly, action-first shape, and keeps the teaching depth. Select it from /config.
Personal Claude Code UI tweaks as a mod: bold numbered prompts, a /prompts pane, a separator after each turn, quiet tool rows, dim narration and framed answers.
Quota-spend projection engine for statuslines. Reads the Claude Code rate-limit payload, logs a rolling per-profile usage sample, and emits a JSON verdict per window (5h/7d): are you on pace to blow the limit before it resets, or coasting under it? Weekly projection counts active hours, not 24/7. Bring your own rendering.
A pane beside the conversation for the work behind it: running shells and agents, the model and a clickable effort meter, and step-by-step progress of any job as a tree: /deck
A pomodoro timer on the prompt's hint line whose break lands while Claude works, with buttons beside it, a history and a report: /pomodoro start
Subagent and Workflow agents created without an explicit model run on Sonnet instead of inheriting Opus; Opus only when asked on purpose