Best Claude Code hooks

A Claude Code plugin that shows 5-hour and 7-day plan-limit bars with reset countdown, burn rate and time-to-empty estimates, session cost, cache hit ratio and a tokens-per-turn sparkline.

A live cactus inbox pane inside Claude Code. Your agent's questions show up as a pane; answer them in place.

Hide API keys and sensitive identifiers before tool results reach the model.

Build, format, and review Power BI reports (PBIR), with a report pane that follows pbir.

A Claude Code Mod that shows your Claude usage as three gradient progress bars right above the prompt in Claude Code Desktop (also VS Code and the mobile app).

Summarize edits, runs and checks as a receipt for each turn, flagging unverified claims.

A Claude Code mod that draws Elizabeth (from Gintama) walking along a context-window progress bar above the prompt, with usage breakdown, turns left to auto-compact, rate limits and session cost.

Draw diagrams, LaTeX math, and tool-result images inline in kitty-graphics terminals

The author nearly wiped a Kubernetes namespace after approving an agent’s `oc delete --all` without reading it, then built a suite of 11 Claude Code plugins using the new function hooks API. The main mod, guardrails, blocks destructive commands like mass kubectl/oc deletes, rm -rf, force-push, and secret reads, and tells Claude the user blocked it. Other mods cover activity monitoring, desktop notifications, context tracking, usage limits, prompt coaching, command discovery, and a mod manager. Nine of eleven never call a model, there is no telemetry, and the project is free and MIT licensed.

Beta, read-only: shows the current /mandrel-deliver Story above the prompt

Mods companion for claude-seo: a spend guard on paid SEO APIs, a live audit band with a receipt, an economy mode, and a visual cockpit with Search Console, rankings, Core Web Vitals, audit, Maps and spend views.

hitokoto: A line from Hitokoto (一言) in a band above the prompt, refreshed on a timer, per session, per prompt or once a day.