Claude Code Model Router Mods

Use ChatGPT, Cursor, Zen, Antigravity and Grok models in one Claude Code session

Routes each task to the smallest sufficient loop of workers

Hand tasks to cheaper models through a local proxy from Claude Code

Rate each prompt with a small model and run the turn at matching effort

Move context compaction off Claude and onto Gemini to cut Claude usage

Difficulty-based routing for Claude Code

Time awareness for coding agents, in Claude Code and Codex: in Claude Code, stamps every message with the local time and how long the turn took; in…

Clawd, a token-saving assistant: trims noisy tool output and pasted logs before they fill your context, tells you when to compact (and does it in one…

Paste a link with a short question and Claude's answer opens with a verdict for your projects: use it, later or skip, the effort, the cost and one…

Your Codex pet, living above the Claude Code prompt: it works while Claude works, waits when you are needed, and celebrates when a turn is done

Claude가 한 일을 전부 일러바치는 Claude Code mod: 상태줄(모델, effort, 요금제 한도, 세션 작업량), 대시보드 패널(/snitch, 에이전트 현황, 변경 파일별 변경 이유, 커밋별 누적, 전체화면이면 우측에 자동 표시)

Buddy is back: the retired /buddy terminal companion as a Claude Code mod, with new sprites and Haiku writing its lines

Makes Claude Code minimal: progress bars per step, code changes in boxes, short answers and a /o model and effort menu

Folds a finished turn's work under a "Worked for" row, leaving the final answer, like Codex

Context-size meter, noisy tool-output pruning and a cost audit for Claude Code, using OpenAI Decisions (gpt-6-luna)

Routes each prompt to the cheapest capable model

Model, effort, context and 5-hour limit under the prompt; outlined prompts and tool rows with icons

Shows which model (Sonnet 5.5, Opus 5.5, Haiku…) each subagent runs on, on its Agent row and in the agent viewer

Shows each spawned subagent's resolved model and effort beside its row, so a surprise model assignment is visible at a glance

Automatic gear shifting for Claude Code: picks the model and effort each message needs, never above your model picker, so your usage limits last…

The MatSci octopus runs around in the band above the Claude Code prompt and reacts to the session

Chat with OpenAI Codex CLI inside Claude Code: Codex answers in the same conversation with native tool rows, diffs and questions, and you switch back…

Makes Claude Code look like the Codex TUI: codex-style tool rows, justified replies, block spacing, footer, and a matching Codex color theme

Picks the reasoning effort for each turn of the main conversation with a System One classifier (hosted Jev or a compatible service) and shows the…

Live status under Claude Code's prompt, and receipts on the turns themselves: context jumps, cache rebuilds and tool failures marked on the turn's…

Claude の応答から Haiku が重要な語句と確認が必要な語句を選び、ピンクとオレンジで描く（Dracula の配色）

Switch a Claude Code session between Claude and a local Lemonade model mid-conversation: /lemonade lists, downloads and switches models

Shows the model in use above the prompt, colored by family: Sonnet in blues, Opus and the latest model in reds

Model router: a steady Sonnet 5.5 main thread does the work; Haiku 5.5 workers take bulk reading and test runs; a read-only Opus 5.5 advisor reviews…

Pick the model and effort in one quiet grey row above the Claude Code prompt

/revo-flow <demande> : l'orchestrateur lit les specs, fait un plan, répartit les tâches entre des sous-agents (modèle et effort choisis pour chacun)…

Tier: the main session manages, helpers do the work

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…

Type in Chinese, talk to the model in English: translates prompts and replies with local Ollama, falling back to Google Translate

Sets the reasoning effort for every prompt automatically, shows it under the prompt box, and counts down how long the prompt cache stays warm

OpenAI Codex as native Claude Code subagents: the Agent tool starts one and the mod drives codex app-server in its place, no model in between

Queues messages sent while Claude is working and sends them one per turn, like Codex

Under each long reply, a box built from the reply's own lines (haiku only picks line numbers): status, outcome, what you have to do, collapsible…

控制台：輸入框上方的 Session／Tool／MCP／流程／記憶／額度 按鈕與側邊面板（未結案／暫時結案／已結案 session 點選切換、工具清單、MCP 伺服器啟用／停用、計畫進度與本 session 即時活動、記憶檔檢視與複製路徑、Claude／Codex…

A Toolbar button under the prompt: pick the model and effort, open the project in VS Code, and switch clean view, which hides the work and shows a…

Codex inside Claude Code: it picks up the chat when your Claude limit runs out, talks with you through Codex's live voice, makes images Claude can…

Tools menu for Claude Code: Clean View, Prompt Upgrade, model/effort control panel, Agent Dock, Roadtrip minigame

Checks your message before it is sent, by rules and, where worthwhile, briefly with Haiku 5.5 (Sonnet 5.5 writes rewritten versions in level auto): a…

A bar under the prompt with compact, clear, effort, model and a color per folder, plus a side chat for quick questions that don't interrupt the task

A one-line TL;DR under Claude's long replies, written by Claude Haiku; the notify mod uses the same line as its turn summary

Git at a glance and in reach: a branch/changes readout, a /git pane to diff, stage, revert and commit, and Conventional Commit messages written by…