Claude Code Mods

Sureffi/week-band

The plan above the prompt: a page's dot block, one node a packet, its class the state, its tooltip the done-when

sebastiandammeyer/claude-mods

Turns Claude's weekly usage limit into a daily budget with a weekend reserve, a status line and a chart

Savo2610/claude-mods

Zeigt, welcher Checkout eines Repos wo läuft und wer daran arbeitet, gibt jedem Worktree einen festen Port, startet und stoppt Dev-Server, räumt gemergte Worktrees auf und schützt Checkouts anderer Sessions.

nickohold/what-what-what

Keeps three lines in a band above the prompt: what Claude is doing right now, for what higher goal, and what it needs from you.

YanZiBin/wheel-finder

Don't reinvent the wheel: finds existing GitHub projects, MCP servers, skills and plugins that fit the conversation and shows them in a small pager above the prompt.

Squidsword/where-did-claude-edit

Shows the branch, worktree and name of the file the session last edited on the hint line under the prompt.

xyzbuilds/while-you-wait

While Claude works, the side panel lists what you can do meanwhile on GitHub: pull requests waiting for your review (with the diff and a quick pre-review), your own pull requests that need you, issues assigned to you.

kikostefanov-lab/claude-code-mods

Claude draws Mermaid, D2 and PlantUML diagrams (UML, flowcharts, sequences, architecture) into a side pane with history, versions, export, copy and share.

natsume-777/claude-mods

A whiteboard Claude writes to: titled Markdown cards (steps, running jobs, links) that stay put instead of scrolling away in the chat

spiiritual/claude-code-stuff

Understand what you ship: plan with a real diagram and decisions you make yourself, then get grilled on it after building.

tornikegomareli/claude-whoami

Claude Code mod: shows which Claude account, organization and plan the session uses and how much of the usage limit is left, and holds prompts in a folder pinned to another account.

kirishimarisano-rgb/Mods-Claude-code-0-powerd-by-claude

Sends each local session's status (what Claude is doing, task progress, context and usage) to the Claude Widget desktop app on this computer, and takes its Stop button

przemonides/winamp-skin

Classic Winamp 2.x skin for Claude Code: main window above the prompt, playlist pane, /winamp to switch back to the classic look

20alexl/windvane

Babysits a long Claude Code session so you don't have to: the context fill is watched, the recorder drafts the checkpoint, compaction happens at the right point with the state banked and the work is resumed, the rules and the checkpoint ride inside the compacted conversation and every subagent's prompt, tool results are trimmed and secrets redacted at the door, project memory and rules are a tool call away, and a token ledger, a pane and a status segment show where the session stands.

20alexl/windvane

Brings windvane back when Claude Code unloads it mid-session, so its tools return without /reload-plugins.

iniphi/witness

Keeps a ledger of what you asked for and counts it done only on evidence: claims stay pale until a read-back confirms them, bare SQL UPDATE or DELETE writes are held for your answer, and a print of what was and was not witnessed goes into the session log.

WizardingCode-io/wizardingcode-developer-crew

Guided delivery where done is proven: the crew engine detects your project, records test-first, runs the real build, tests and checks, and holds the turn until the evidence says the work is done.

StarHawkMK1/cc-mods

A one-line band above the prompt: the stage and progress of a superpowers flow, another skill or Claude's own to-do list; the last test run; the request's time; and how many subagents are running. Comes back on resume.

ilikeeatingrice/claude-code-work-panel

A right-side Claude Code panel for your task tracker: browse roadmaps, start a ready task, and clear-and-start with a handoff note when you stay in the same roadmap

eascendco/workbench

Your current task above the prompt and a Workbench side pane: the task and its work items, what Claude is reading and editing, git status, file search and a file tree. Tasks are plain markdown files in plans/.

fanbuz/claude-workbench-mod

Workbench for Claude Code: a live narration band above the prompt and a side pane with this turn's steps, working tree changes and turn history.

harshitmywork17/claude-mods

One workspace for watching and understanding Claude's work: a one-line band above the prompt and a pane with Now, Changes (file diffstat, changed functions, highlighted diffs, replay), live Preview, Artifacts, Code Map and Usage. /wb

Pigula1984/workbench

A set of mods that put live session data around the prompt box: statusbar (model, effort, mode, directory, branch, usage) and workflow (a dynamic workflow's steps under the prompt).

lokeshrana9999/workboard

Live view of long-running work: a /board pane with a flow diagram and summary, a band above the prompt while subagents run, and a status tool Claude can call.

VictorGjn/workflow-gate

workflow-gate in the Claude Code terminal: the plan as a flowchart with editable tiers and missions, an Opus review before you approve, the run live card by card (progress, cost, each agent's story), reasoning checks with an Opus orchestrator, retros, and learning across runs. Requires workflow-gate.

guojunjie0614-cyber/workhorse

Hand coding, review and ops jobs to Codex as background subagents, with a live status band, watchdog warnings and completion notices

cprentice9/ccmods

While Claude works, Clawd walks above the prompt: across a starry sky in the desktop app, Ghostty and kitty, stopping to look around, wave and jump, and back and forth in half blocks in other terminals.

santtisosa/worklog

Time you actively work with Claude, per repo and branch (or ticket), with a weekly view and a CSV timesheet.

KeygripGit/claude-mods

Keygrip work orders in a pane: /workorders lists your project's deploy queue; pick one to read its fields, edits, schema and verify steps, see the page before and after the edits — a page that is not live yet is shown on a template page of your site — and approve it for publishing; the queue marks each page approved, awaiting approval or live. Paste your project's client key once. From the agent's first Keygrip tool call on, the session's conversation is sent to Keygrip to improve the next page.

atsusta/workspace-guard-mod

Blocks reading .gitignore and creating venvs or heavy media inside the Syncthing-synced claude-code workspace.

EggmanPDX/claude-mods

Enforces Egg's CLAUDE.md hard rules on Bash: account isolation, push protocol, Vercel D8 prebuilt, no --no-verify.

srwiseman/claude-mods

A live pane for a workstreams round: each stack layer's place in the review loop, its CI, threads, agents and collisions

harshshah-2023/Claude-Mods

A persistent file tree and timeline of the files Claude reads, searches and changes, with summaries and exact diffs.

thieung/claude-mods

Guards a shared worktree: stops `git add -A` from staging files this session did not touch, previews destructive git commands, and points at the process holding a busy port.

cmorss/claude-mods

/merge-status: a pane with what merged into local main and which git worktrees conflict with it, and a link to each worktree's session (Claude desktop app)

floheissler/cc-worktree-radar

A live radar of your parallel branches and worktrees above the prompt: which merge cleanly, which conflict, which are stacked, which a Claude session is still working on, and the order to land them. Read-only: it never merges, rebases or moves a ref.

netgfx/claude-worktrees-mod

A /worktrees side panel: colored, named worktrees with one-line tasks, fast session switching, and per-worktree dev-server ports Claude respects

lisim/wrap-buttons

After a wrap-up reply, shows buttons to /clear and carry the handoff's next step into a fresh session.

AliakbarNateghi/cc-wt-gc

/wt-gc lists git worktrees with their PR/merge status; /wt-gc prune removes clean merged or closed ones.

orangeJigglypuff/better-btw

Select text, run /wtf, and a side pane explains what it means in the context of this conversation, without adding anything to the conversation.

gawlik81/claude-mody

Wytłumacz: po polsku i prostym językiem wyjaśnia krok po kroku, co Claude właśnie zrobił (na bazie explain-it)

dvakatsiienko/frame

reads every Bash call, fork spawn and Write before it runs: a Write over a tracked file the session never read, a floor command, a system, overwrite, prune or remote-delete command, a known hazard shape, a coder spawn whose brief has no x brief check stamp, or a fork with no why-fork line is refused with its safe door, never a dialog; each refusal and escape is kept in its own store as an event and a per-day count, which x-mod-stash's band reads as its counter row

48X-Dev/claude-plugins

Saves Claude Code's 5-hour and weekly limits, context fill and session cost for the 48X AI Usage app.

narrowstacks/claude-code-mods

In Xcode projects only: one line with the last build and test result and the booted simulator; flags 0-test runs; blocks a second xcodebuild while one runs

iyiob96/Context-Mod-Companion

Context-window band above the prompt with RTK token savings and a clickable pixel Yemeni companion whose mood follows how much context is left

davidtheproduct/your-move

Find your messages. Never miss what Claude is waiting on you for. Borders your own messages and outlines any part of Claude's reply that needs you to act. Claude Code in the terminal only.

cdeust/claude-mods

Automatic effort and model, and a quota that limits how many agents run, not how well one runs. Every request runs at the effort its grade calls for (zetetic-genius) and the main tool loop steps down the ladder. A subagent runs on the model its Agent call names, else the one its agent definition pins (read from the definition file, since agent.spawn carries only the call's), else the one the DESIGN block of its prompt names, else, for an inheriting type nobody designed, the rung of the dispatch table (hooks/dispatch-table.ts) for its task's class: one haiku completion grades the spawn prompt (routine, planned, bugfix, analysis, critical; a review or not), and the table gives model and effort (no haiku rung; a review runs on opus when complex, else on sonnet). A classifier that gives no class decides nothing: the spawn inherits and a POLICY row says why. Its effort is the Agent call's, else the one its definition sets, else DESIGN's, else the table's; an inheriting type the call gave a model runs every step at the effort of the turn's grade (main's spawns on a graded turn; a fork, a workflow agent or a teammate is never rewritten). No pressure, quota or context changes a model or an effort the orchestrator designed. Each spawn's requested, definition and decided model, what ran, the effort decided for its steps, the table's class and a short reason for each are kept per session. Quota pressure limits concurrency: the cap on subagents running at once falls from max_parallel_agents (5) at an empty five-hour window to 1 at quota_pressure_percent, and a spawn past it is refused with a message the orchestrator reads (wait for a running agent to finish). Only the main loop's own spawns, a fork of main included, are counted and refused: a subagent's nested spawn, a workflow script's agent() and a teammate are not. Main itself drops to sonnet under quota pressure, and doing implementation work in the main loop is warned about (POLICY row, toast), never blocked. A long streaming tool result is written whole to a file in the session's tool-results directory and the model reads a preview and the path (nothing is cut). Owns `phases`: the phase each loop is in (planning, analysis, design, implementation, verification, test, review, report), read off its tool calls or, for an agent, its type; on each run, the steps the stuck escalation raised (counted where each raised step landed, under enforce only); and the advisor calls of main with when they came. The tool loop of a phase the insights calibrated runs at that rung instead of medium (never above the request's grade). Starts in enforce; observe logs the decisions without applying them.

cdeust/claude-mods

The learning loop: every main turn is kept in a ledger across sessions (phase, effort, steps, failed tool calls, agents, advisor calls), joined with the outcomes observed at tool.call (commits kept or reverted, test exits, PR verdicts and merges, no transcript) and the ones /insights writes, and the project's goals; probe sessions are counted and kept out; a phase whose evidence, read on the rung each run had and with Wilson intervals, says so gets its tool-loop effort moved one notch (every applied move is measured pre/post on the same phase, the same task classes and the steps its ceiling decided), written to a calibration file the autopilot reads. /calibrate apply, /calibrate reset. Starts in observe: proposals are shown, not applied.