Claude
Tools

Loop-discipline patterns for Claude Code: Stop hooks with evidence files, three skills mapped to common failure modes, and reference samples for structured shell, structured tests, and prompt caching.

Tidy up your Claude Code environment — sessions, permissions, context, hooks, schedule. Open-source TUI/CLI.

Methodology and tooling for auditing your own Claude Code session logs. Schema, ingest pipeline, analyses, and LLM-assisted verification-debt audit.

Drop-in admin activity log feature for Next.js + Supabase dashboards. Append-only audit trail of who-did-what, packaged as a Claude Code skill.

Email design skill for Claude Code. Anti-slop emails — six committed archetypes, brand-aware briefs, real imagery, real footers. Works with any ESP via MCP (Nitrosend, Klaviyo, Mailchimp, ActiveCampaign).

Claude Code plugin: extract scene-change frames + audio transcript from any video so Claude can 'watch' it.

A pi-mono extension that ports Claude Code's security and permission system — intercepts every tool call with a multi-stage decision pipeline.

Generate production-ready cinematic video prompts for Higgsfield AI from reference images and project briefs.

Browser Companion is a Chrome MV3 extension scaffold for a chat-first browser agent. The extension observes the active page through constrained tools, keeps risky browser actions behind policy checks and confirmations, and is designed to connect to local subscription-based CLI providers without API keys.

Claude Code skill: tiered (XS/M/L) development workflow with RFC-gating, smoke tests, multi-agent code review, and incremental security levels.

Claude Code skill: bootstrap a new project in one conversation with base templates, opt-in stack extensions, and a self-augmenting feedback loop.

Claude Code skill: turn any visual identity into a paired DESIGN.md + design-showcase.html. Built on the awesome-design-md format.

Open Pets — chibi pet atlases + reusable AI Skill for Codex CLI / Claude Code

🎬 AI Video Creator – Open-source video automation toolkit. Tell Claude Code what you want, get an MP4 back. No accounts, no API keys, fully local. Multilingual helper.

Claude/Codex runtime compatibility gateway. Beyond routing: protocol conversion (Responses↔Chat↔Anthropic Messages), tool-call JSON repair & fake-tool detection, secret redaction, MCP/Shell/Patch safety gates, context compression, agent recovery, provider capability profiling & benchmarks. 第三方模型无缝注入 Claude 和 Codex

Dynamic Claude model + effort routing plugin for Claude Code. 5 tiered subagents, hook-based classification, 16/16 tests.

不動産鑑定士が公開する白箱方式の土地査定OSS。MLIT取引データからヘドニック回帰係数を都度推定、業者用シート（係数全開示）と顧客用シートを生成。Claude Coworkerシリーズ。White-box AVM skill for Japanese land price assessment. Re-estimates hedonic coefficients on every run from MLIT transaction data, generating broker-side worksheets (full coefficient disclosure) and client-facing assessment reports (xlsx). Part of the Clau

A Claude Code skill that lets the agent live-debug any Node.js service via Chrome DevTools Protocol — discovers .vscode/launch.json automatically, sets breakpoints, captures pauses, inspects state, and reports back conversationally.

Drop-in Claude Code skill bundle for AI-resistant SEO article generation — research → write → E-E-A-T audit, configurable for any site or language.

Production AI Operations Suite — 7 MCPs + 13 skills + 6 hooks + 5 monitors. Catches dangerous bash, cost spikes, agent-config attacks, output-action divergence, silent cron failures, supply-chain risks, and unsafe upgrades. Operator cockpit for production AI.

Starter kit for building Claude Code agents, skills, hooks, and scripts

Unofficial Model Context Protocol server for Plunk. 84 tools spanning email, contacts, campaigns, segments, workflows, and analytics.

AI agent code skill to generate valid BPMN XML files.

Claude Code 中文汉化工具 - 工具提示钩子 + 界面汉化 | Based on cute-claude-hooks