Productivity plugins

Connect Claude Code to your own Whisper Money account and its documentation.

Shows suggested next prompts above the prompt box when you press What next?. Type /auto to let it send the first suggestion for you after a countdown you set. Off by default.

A live file explorer panel beside the chat: your project's file tree with git status, diffs and line counts, what Claude reads, writes and commits, search, file previews and your editor one key away. A Claude Code mod for productivity.

Private memory across your agents

Chips above the prompt for each skill, subagent, MCP server, permission decision, hook and rule that activated this session. Press one to see why.

alt+g opens lazygit in a popup over Claude Code; ESC closes it.

Switch model, effort and ultracode with one keystroke; shows the model and effort in the footer.

Demo: colored pills above the prompt: context tokens, 5h/7d limits against time elapsed, prompt cache time left, session cost and running subagents.

See your 5h/7d subscription limits above the prompt, with a burn-pace forecast, threshold toasts and a cold-cache warning.

Keeps the goal, your prompts, pinned notes and touched files across compaction, and archives the history before it.

Cuts 'You're absolutely right!' and other tics: a tone rule, opener stripping and a tic counter.

D&D DM plugin for ChatGPT, Antigravity and Codex with durable campaign state and journal-backed memory

Drives your Claude Code sessions on its own: asks you pointed questions first when a prompt leaves the goal, constraints or check unclear, rewrites a session's first prompt, decides permission asks from the whole conversation and Claude's questions once you've gone quiet, and after each turn sends the practical next step (fix, verify, commit on a branch) until only you can decide what is left, the push and the PR included.

Git-versioned agent memory — capture, compress, and recall session context across agents and machines without re-reading state files every boot.

Audit global Claude Code installs against real usage, turn them into per-project cc-* kits, and keep your global setup lean.

While Claude works, opens the windows you chose (a feed, a film, a PDF) and minimizes them when Claude needs you or is done.

Notification sounds for Claude Code: when Claude finishes, asks you something, needs a permission, runs tests or hits a limit. Presets of your own are folders of .wav files.

See every Claude Code session on your computer, open one, move it (or a whole project) to another folder, and delete sessions or the orphans, always with a preview and a confirmation.

An OpenCode-style look for the Claude Code transcript, with framed shell commands; every color configurable.

A register of the shell commands Claude gives: /cc opens a panel to browse them message by message and copy one.

A living board per session: Claude keeps the list of what is done, pending and waiting on your decision with a board tool, and /board rewrites it from the conversation.

Branch Canvas: local conversation trees and branch navigation for Codex. Copyright Rui Li; all rights reserved.

No description.

No description.

No description.

Live usage meter under the prompt: context fill, rate-limit windows, session cost and token totals

총괄 상황판(읽기 전용 모드) — /forge-board 가 열린 PR · 이 PC 일감 · 관리 가드 상태 · 팀장 방을 패널에 보여 준다. 도구 호출 훅 없음 — 차단 훅을 건드리지 않는다(forge#4191).

Turn a braindump into tomorrow's runlist. Input: unstructured text + optional previous runlist. Output: 4-quadrant Obsidian markdown (~800 tokens) with entry points on high-activation items, carry-forward tags, kill reasoning, and RUNLET_SUMMARY JSON for agent pings. ND-adapted: classifies by activation energy, not urgency.

Claude Code status line with the model, effort level, context window usage, and whether caffeinate keeps the Mac awake.

Work with ProcurementExpress from your agent: pending approvals, budgets, purchase requests, supplier documents and invoice-to-PO matching.