
The vibe-coding seatbelt for Claude Code — catches AI coding mistakes · 바이브코딩 안전벨트, AI 코딩 실수를 막는 플러그인
2,704 developers. Ranked by the GitHub stars their repos have earned. Each card shows their biggest plugin.
2,704 developers

The vibe-coding seatbelt for Claude Code — catches AI coding mistakes · 바이브코딩 안전벨트, AI 코딩 실수를 막는 플러그인

Before you accept an agent's diff, see what changed in behavior, flow, contracts, and architecture — with every claim grounded in real code. ▶ See live examples → · static cards from a fictional storefront, exactly what Mappamind renders for your repo

CSW takes a ticket from your tracker to a merged pull request, then cleans up after itself. Superpowers owns how the work gets done. CSW owns *how it gets shipped and closed out*. This is one person's idiosyncratic workflow, and it is probably not yours.

A growing collection of GTM skills for Claude Code, built by 5050Growth. Skills you can drop into Claude Code (or run as plain Python) to handle the GTM work that doesn't justify a SaaS subscription.

Build Zalo Mini Apps - lightweight web apps running inside the Zalo super-app platform.

Portable, install-once skills for AI coding agents. Each skill is a single SKILL.md of plain-markdown instructions your agent loads automatically the moment your request matches — no prompt to paste, no copy-paste drift, the same discipline every time, in

A Claude Code plugin that enables powerful data transformation and export using DuckDB SQL.

A single, language- and domain-agnostic engineering discipline you drop into your AI coding flow (Claude Code, Cursor, or any agent that reads skills / rules files) so it builds like a senior engineer instead of vibe-coding a demo that breaks on contact with

Hold a natural spoken conversation with a coding agent — Claude Code or OpenAI's Codex — while it does real agentic work. Your Mac runs everything — the audio pipeline, the sessions, the state.

Your docs can never lie again. You change code and the README quietly stops being true. Nobody notices until a stranger does.

Audits what your Claude Code agent did against what it said — reads session logs and git, cites evidence, calls no model

Brand-agnostic design system and artifact generators for Claude Code, plus a figure-worth audit and PDF and HTML-to-PNG export utilities. 19 skills total. Install via the slogsdon-claude-code-config marketplace.

One small Go binary that lets an agent inspect a remote server read-only, without copy-pasting logs and configs by hand. rpeek serve — the server. Copied onto the remote host and run there.

Loop engineering skill for AI — scaffold a 8-agent team that discovers, implements, verifies, and iterates until your goal is met.

A local-first personal voice layer for AI agents. Bring your voice to any agent. Adapt to any audience without losing yourself. Humanize AI-generated drafts by rewriting them in your own calibrated voice: PR comments, Slack replies, docs, RFCs, and emails.

A Claude Code harness that keeps sessions small and smart: a context line that orders the handoff before a session gets expensive, Sonnet agents for bounded work, a short task-ledger skill and a usage check. Built at Flowy and measured on real sessions there.