Prompt
593 mods. Bands above the prompt and tools that rewrite what you send.
- je-coupe 0.1.0 End-of-session ritual: typing 'je coupe' asks Claude to update the doc files and shows a band of the ones still untouched 0
- lgtm-flash 1.2.2 After every code change, the LGTM Lines container ship sails in, the L G T M containers slam onto its deck, and the camera zooms in on LOOKS GOOD TO ME 0
- villager-hp 2.5.2 Your Claude session and weekly usage limits as an animated villager health card, with rank badges and a death screen at 0 HP 0
- live-progress 0.1.1 Live progress bands above the prompt for long-running work Claude starts: tests, builds, scripts, logs 0
- skills 0.1.0 /skills: per-prompt skill suggestions, what Claude actually loaded, and one-key use of a suggested skill 0
- clowl 0.1.0 Clowl: eine treudoof schauende Eule über der Eingabe, solange Claude auf die nächste Aufgabe wartet. 0
- status-band 0.2.1 Status band (above the prompt): model, working state, context, cost, branch, effort and rate limits 0
- agent-crew 0.1.0 Bubbles above the prompt showing each subagent as a mini Claude, colored by what it is doing, with a progress % and a short label. 0
- memory-bar 0.1.0 Shows how full Claude's context window (its working memory) is, as a bar above the prompt. 0
- task-steps 0.1.0 Shows the steps Claude plans for your task, and its progress through them, above the prompt. 0
- modbar 0.1.0 Draws the area above the prompt for other mods: their blocks, packed into configurable rows 0
- handoff 0.1.0 A note per project folder for the next session: /handoff writes a full one, and with /handoff auto on exit saves a quick one when work happened; the next session gives Claude the note 0
- context-budget 0.2.0 Compact reminders as the context grows, a check before resuming a large session after the prompt cache expires, and per-type models with a minimum effort for subagents. 0
- effort-gate Denies a subagent or background session running Opus at xhigh or max effort unless the user said so 0
- context-vigil-mod 0.0.0 Test harness only: lets claude plugin test run tests/ against plugin/. Never installed; the marketplace points at plugin/. 0
- session-progress 0.1.4 Animated progress band for long-running turns, with a completion summary 0
- pocketpet 0.1.0 A tamagotchi-style virtual pet above the prompt: feed, play, clean, medicine, lights, discipline 0
- close-gate 0.1.4 Show session close facts and give them to a handoff. 0
- focus-panel 0.2.0 After each response, a panel above the prompt with what needs you, errors, the next action (with its task id), open to-dos and heads-ups. 0
- prompt-cleanup 0.1.0 Fixes typos in what you type before Claude reads it 0
- hold-em 0.1.0 A six-player Texas Hold'em Sit & Go against five bots, played in a four-row band above the prompt. 0
- auto-model 0.5.0 프롬프트 캐시를 아끼며 모델을 골라요. 한참 쉬어 캐시가 식은 긴 대화엔 /compact 를 제안하고, 가벼운 요청은 바꾸는 쪽이 확실히 쌀 때만 Sonnet 으로 내리고 검색·탐색 서브에이전트는 Haiku 로 보냅니다. 테리 카드에 표시 0
- agent-bar 0.1.0 Keeps the savvy-progress bar on for every task: the todo list, tasks and every agent the main thread starts become rows, counted done as they finish, unless a skill is already reporting 0
- polish 0.1.0 Cleans up dictated prompts with Haiku: drops the ums, false starts and spoken punctuation, and spells your jargon right. Learns your vocabulary as you type. 0
- next-steps 0.1.0 After each reply, shows suggested next steps as buttons above the prompt (press 1-4 or click). 0
- model-router 0.1.0 Classifies each turn as mechanical, ordinary or hard and routes it to a matching model and reasoning effort, logging every decision. 0
- context-handoff 0.2.1 Context 用到七成時,在輸入框上方顯示交接按鈕:寫交接摘要、存檔、清空對話,再把摘要當新對話第一句送出。 0
- task-progress 0.1.0 A progress row above the prompt for each plan the model reports, plus a row per running subagent. The bar shows what the model reports; it costs no tokens to draw. 0
- assumption-tracker 0.1.0 Claude logs the assumptions it makes mid-build; a band above the prompt expands into a list where you mark each Right or Wrong. 0
- idea-shelf 0.1.0 Shelve side ideas while Claude works, then send them with one click when you are ready. 0