Prompt

Skills and guard hooks: ai-code-cleanup, interrogate, shape-task, verify-change, setup, a shared-worktree git guard, and a guard against reading API-key config files

Cross-model review and delegation: /mm:review and /mm:run, built on mmrun

Next-prompt prediction: after each main-thread turn, a fork of the session's own model predicts your next message and shows it as the prompt box's suggestion, Tab to take; /prediction-stats reports how often it was taken

A per-project progress board: after each typed turn that edited files or made a commit, the plugin records its work as nodes through a Sonnet completion, kept under ~/.claude/progress/ or committed with the project; the plugin mirrors it to an optional Artifact canvas

Painel ao vivo do que o Claude e cada ajudante fazem: subagentes, agentes de Workflow, comandos, agy e Codex em árvore, com modelo, tokens, pendências, avisos, outras sessões e uso

Depois de cada resposta, o Haiku propõe três próximos pedidos em botões acima da caixa de mensagem (teclas 1-3)

A side pane that shows the agents working on your requests as illustrated characters

Thumbnails of the images you paste, above the prompt: the real picture in Ghostty and kitty, colored cells in other terminals, and the real file behind a copied Finder icon

A Tamagotchi-style coding companion: it earns XP from edits, reads, passing tests and commits, levels up and evolves, and needs feeding, play and rest. /codepet

A live pane: context ring, session cost, a colored timeline of every tool call, files edited, and a sports-commentator recap of each turn

A workbench for building Claude Code mods: which mods loaded, a live tape of every event with its real payload, the four ship gates on the mod in progress, and tools so Claude can read payloads and run the gates itself

Past a context threshold, records a hand-off and continues the work from the notes: in a new desktop-app session by default, or in this one after a compaction or the app's clear

See what is eating your Claude Code context, live in a side pane, and fix it in one key: top context eaters, prompt-cache countdown, compaction that keeps your decisions, handoff notes

A band above the prompt: context against the auto-compact window and turns left to it, prompt cache countdown, rate limits with their pace, a cache hit warning, cost, and stand-up / late-night reminders, fitted to the width. /meter shows the details.

A live todo pane for the session: the agent keeps it current through a tool, you read and tick it without asking.

Nexika mizan: keeps your Claude Code session in balance. A band above the prompt shows the branch and who started it, open PRs or MRs per person, the branch's CI (passed, failed with the job, running), RAM and disk with warnings, the context level (fresh, mid, full), the running agent and its task, the current task step and the session cost; /mizan opens the details. At full context it saves a...

Cleans up after merged work, once it shipped: removing a branch needs git, the forge or the user to say it merged; removing it or filling in its issue waits for the pipeline holding it to pass; merged, clean work gets a follow-up to remove its worktree and branches.

📸 Shows you the pictures: a thumbnail the moment you paste one, and /gallery of every image you and Claude passed around this session.

Shows pasted images as framed thumbnails under the prompt; click to expand, open in a pane, or /image

A pane beside the transcript: the questions you asked (jump to where each was asked and answered) and the steps the model set itself, with completion rings

Pasted images become chips you can click to open in the system viewer, in the transcript and above the prompt

Band tracking session tokens, tool calls, model working time and cache countdown, with a pane that shows each subagent's transcript

Banda sobre el prompt con el contexto (cifra y barra de verde a rojo), el gasto de la sesión, el uso de CPU y RAM del equipo y el límite semanal del plan; hace que el motor compacte, también a mitad de una tarea, al llegar el contexto al umbral, que se cambia arrastrando la marca de la barra, y guarda el resumen de cada compactación. Panel lateral con los subagentes de cada tarea (inicio, fin...

A lamp above the prompt: red when the firstmate fleet needs you, green when a PR is ready.