Claude Code Mods

rotembab/jarvis-claude-mod

A JARVIS-style voice for Claude Code: say "Hey Jarvis" (or hold push-to-talk) to speak, hear the reply, talk over it to interrupt. Local wake word and speech-to-text, Fish Audio speech, optional webcam hand control for the mouse and windows, control of your TVs and smart home devices, and PC control: desktop actions such as opening apps and the media keys, with a guard that asks before Claude's risky commands.

JayDoubleu/cc-mod-jev

Jev-scored context pruning for Claude Code through OpenRouter: at compaction every tool call and result is scored by TypeSafe's Jev decision model, stale ones are dropped or truncated, everything kept stays verbatim. An optional gate truncates oversized tool outputs as they arrive.

roma-vibe/jev-governor

Jev-guided model, effort and subagent routing for Claude Code (Opus 5.5 / Sonnet 5.5), auto-created specialist subagents and Jev-pruned compaction.

DmitryBMsk/thinkdial

Picks the model and reasoning effort for each task with TypeSafe's Jev. Routes subagent models at agent.spawn and subagent and main-loop effort at turn.step, with optional main-loop model routing. Falls back to the engine's built-in classifier when no key is set.

Kiy-K/oxide

Picks the model and the reasoning effort for each task with TypeSafe's Jev, a System One decision model reached either through TypeSafe's own API or the Vercel AI Gateway. Routes subagent models at agent.spawn and the main loop's effort and model at turn.step, in both directions, falling back to the engine's built-in classifier when no key is set.

drewpayment/jev-route

Routes each turn to the cheapest capable Claude model using Jev, TypeSafe's decision model. Toggle with /route.

Hoppin-Sorter/jev-router

Asks Jev before each prompt what the task is and how hard it is, then picks the Claude model (and optionally effort) and the skill that fit. /jev and the Jev Bar menu bar widget set the quality/cost focus.

karanb192/jev-skill-scout

Before each prompt reaches the model, TypeSafe's Jev ranks your installed skills against it and attaches one line naming the skill to load. Sends the prompt text to api.typesafe.ai. Needs function hooks (early access).

cardosoads/jev-suite

Branch git acima do prompt e juízos do Jev (TypeSafe): guarda de comandos, roteador de esforço, detector de segredos, aviso de agente travado, dica de qualidade do prompt e triagem de saídas

GastonGelhorn/jevmate

A fast decision layer for coding agents. Jev ranks files, picks the tests a change can break, sorts failures and reviews diffs by risk, so Claude reads only what matters and judges only what Jev is unsure of. Also: long output trimmed, compaction that keeps what the work still needs and saves the rest on disk, safety signals on Bash and fetched content, and the session's numbers above the prompt. Thresholds measured, not guessed.

unclecode/jevshift

Experimental Jev-assisted model and effort selection for Claude Code: observe, pin, off, and opt-in auto.

dodofk/jojo-sbr-token-usage

Context window usage as one line above the prompt, told as a Steel Ball Run race: Gyro's five lessons and a horse running to the finish flag.

KTCrisis/flux7-mods

Plays music from YouTube when asked in plain words, audio only through a hidden VLC, with a small control pane

Hnquangg123/kessoku-ultimate-bocchi

Kessoku Ultimate Bocchi Theme: a Bocchi the Rock!-inspired charcoal-and-pink skin for all of Claude Code (an engine theme plus every render site), a live Kessoku Band panel with the band on stage as animated pixel-art sprites under spotlights and their subagents as songs on a setlist (solo any member) and band subagent types Claude can delegate to by name, a recap card when a long set ends and a per-project tour across sessions (/bocchi stats), and themed Cache Tax, Blast Radius, Filetree, Reflect and Replay Theater gags.

YanZiBin/knowledge-card

While the AI works, a plain-language knowledge card above the prompt explains a concept from the current task. Click 我知道啦 for the next one.

tomshaw/claude-code-plugins

Runs Pint and Larastan after Claude edits PHP, and Pest or PHPUnit on demand, then shows the results in a live pane and the status line. /laravel-tooling opens the pane, /laravel-tooling run checks everything, /laravel-tooling tests [filter] runs the tests.

pepperonas/last-push

When this project was last committed and pushed, in one line under the Claude Code prompt: unpushed commits, uncommitted files, staleness colour and a reminder

arthgirard/claude-latex-render

Renders LaTeX in Claude's replies: display math and TikZ figures as crisp images (kitty graphics protocol), inline math as Unicode

chang-07/cc-latex

Draws display LaTeX ($$...$$ or \[...\]) in assistant replies as typeset images (tectonic + kitty graphics)

paragpandyareal/lazy-panda-panel

Review docs without lifting a paw. Your files open right where you’re working. Excel, Word, PDF, Markdown, Confluence (ADF), HTML and pictures: highlight a part, comment, and Claude edits the real file. Nothing to install (Word, Excel and PDF files over 4 MB need Python).

LeanZo/lean-switchboard

Lean Switchboard: one list of every plugin, connector and skill in Claude Code, each with an on/off switch for new sessions

BuddyLim/claude-code-mod-lens

Review a repo's changes in a pane: the changed files with editor-style diagnostics (pyright, ruff, tsc, eslint, terraform), a diff view, a git graph, commits and stashes, and pull request comments. Run /lens [folder] [base].

PedroLaRosa/claude-code

Puts tsc, linter and secret-scan diagnostics in the result of every Edit and Write, so Claude fixes what it broke in the same turn

quango2304/leo_claude_mod

Leo's Claude Code mods: usage bars, limits, cost, prompt cache countdown, active time and running agents

burnrate-ai/burnrate

Warns when the prompt cache is invalidated; near the 5h cap blocks subagents and wide scans and asks Claude for short answers.

VaitaR/claude-code-limits

Claude Code mod: your 5h/7d quota, context window, prompt-cache time left and session cost on one line above the prompt, with per-subagent cost on hover

burnrate-ai/burnrate

Band above the prompt: 5h/7d limits with reset timers, context fill, cache hit rate and what fills your context.

mberto10/applied-systems

A Claude Code mod: a Linear project as a tree and a flowchart in a side pane, with progress, search, filters, shortcuts and markers on the issues agents are working on.

Sarimarcus/claude-code-plugins

Linear workflow for Claude Code: issue-next, issue-start, issue-review, issue-ship and issue-plan-cycle skills backed by a deterministic CLI, a linear-manager agent, and a mod that shows the current issue above the prompt with your other sessions and conflict warnings

tripolskiydigital/claude-links

A links bar above the prompt in Claude Desktop's Code tab: up to 4 pinned links per project or session with favicons, the session's 10 latest links, and a pane of every pinned link. Reads the session transcript; asks a link's own site for its page title and favicon (curl for PNG and ICO icons); opens links with open.

juninmd/cc-litellm

See your LiteLLM virtual key inside Claude Code: budget, spend, limits, expiry, models and recent usage. Admins can also create and edit keys, grant extra budget and read router fallbacks.

benjaminmodayil/live-recap

A card above the prompt that says what Claude is working on, refreshed every few minutes while a turn runs

connectwithub/loopkeeper

See and cap your subagents: cards with cost, tokens, context and progress, spawn caps with an override you approve, tree and team views, idle alerts and a Stop button.

connectwithub/loopkeeper

Stop runaway token spend: budgets per session and per subagent with warn, deny-agents or stop actions, plus a prompt-cache countdown, cold-cache pings and keep-warm.

pepperonas/loc-today

Today's lines of code by Claude, across all sessions, under the prompt: new, edited and deleted lines, and on hover the split by project, largest first

StartupBros-com/hov-marketplace

A second pair of eyes while Claude works. Every few model steps it asks one tool-less fork of the session (served from the prompt cache) whether there is something you should know and might miss, and shows the answer above the prompt. No model call while nothing draws the session.

StartupBros-com/hov-marketplace

Stops Claude repeating the same failing tool call. Counts identical calls that error or are refused, hints after the second, refuses the next after the threshold with a reason Claude can act on, and shows a worsening face above the prompt with a one-key Stop and rethink.

MiguelProz/claude-loose-ends

Proposes the loose ends Claude leaves undone and lets you decide which to keep; keeps them in refs/loose-ends without touching your branches, and Chispa flags what needs your attention. English and Spanish.

lumberroom/lumberroom-claude-code

lumberroom as Claude Code's memory: the digest in the system prompt, a periodic search-and-write reminder, optional per-prompt recall, built-in memory off.

luancamara/luna-pruner

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

miiraheart/marginalia

Select text in the conversation, then quote it into your prompt (it stays outlined), ask about it, turn it into a task or keep it. Theme mods can draw its bar in their own style.

clockendbader/mascot-studio

A 2010s animation studio in your terminal: Clawd acts out what Claude is doing, with usage and music tabs and Windows 7, Mac OS X and Ubuntu themes.

davekiss/md

Read, write and review markdown with Claude in a pane inside Claude Code: drafts stream in, comments stay on their words, and Claude suggests edits you accept or reject.

DenisGuiraudet/claude-mods

Claude's replies as pixel bubbles: a short sky and hill on top, a strip of grass under the last block

njp-coder/meanwhile

A small card above the prompt teaches you one thing while Claude works, and turns your day into something worth sharing

youngOman/pill-mods

Draws mermaid blocks in Claude's replies as pictures right in the conversation: navy nodes with light-blue borders on a transparent background, download and copy-image buttons.

ejklock/claude-mermaid-render

Renders the Mermaid diagrams Claude Code writes or reads as elegant cards in the transcript: colored Unicode art in any terminal, real SVG on desktop.

benjaminmodayil/live-recap

Prefixes each Claude reply in the transcript with the local time it arrived, with no model calls