AI project showcase / idea exchange / co-creation

Built ideas,made to be seen.

Share your AI projects, browse fresh ideas, and talk through the parts that are almost there. Working demos deserve an audience, and blockers do not have to stay private.

Share projectBrowse co-creation
Share work
Showcase
Find sparks
Ideas
Keep talking
Co-create

Project square

Scan active directions first, then refine by category or momentum.
All projectsShowcaseCo-create
Popular categories
All categoriesAI AppsAI AgentsDeveloper ToolsProductivityDesign & CreativeGamesTools
Sort by
NewestHottestMost likedMost savedMost discussed
Finder Actions
Showcase
Online
Developer Toolscodex

Finder Actions

macOS Finder 原生右键扩展：右键一键在 Alacritty / VS Code 打开目录，支持自定义脚本，运维纯口述让 Codex 半天搓出。

Finder Actions 是一款 macOS Finder 原生右键扩展。在任意文件/文件夹上右键，即可一键在 Alacritty 或 VS Code 中打开当前目录；还支持自定义 shell 脚本动作（路径以 $1 传入）、拖动排序、生效目录控制、中英文切换。 作者 qvgz 本人是运维，不懂 Swift。他的原话：「试一试让 Codex 写，纯口述需求没看一行代码，半天做得有模有样。」项目用 Swift + SwiftUI 实现（作者 GitHub topics 标注 swift/swiftui/developer-tools），MIT 开源，Release 页可直接下载安装。 诚实注记：安装需 Ad-hoc 签名去隔离并在系统设置中启用 Finder 扩展，README 有完整说明；3 stars 的新工具，仍在持续迭代。

V@VIBEFIX
0
0
2
0
Neuro Symbolic Game Level Designer
Showcase
Online
Developer Toolsibm-bob-20claude-code

Neuro Symbolic Game Level Designer

Describe a level in one sentence: agents plan, algorithms build, validators check — a playable Flame level in about a minute.

Neuro Symbolic Game Level Designer is a hybrid AI level-design tool built for an IBM Bob 2.0 hackathon. You describe a level in one sentence; a pipeline of agents plans it while classical algorithms build and verify it. The output is a complete, playable 2D isometric Flame game level — a Tiled map plus a typed Dart loader — with a six-check validation report covering spawn reachability, item placement and pathfinding. The author, raf-xtgt, describes the build on dev.to: "One orchestrator session wrote the architecture and a precise prompt for each task. A coding agent implemented each task, and I reviewed it with tests and browser checks. IBM Bob built the first five tasks… After my Bob credits ran out, I continued with Claude Code using the same workflow." Stack: Flutter web / Flame 1.38 frontend, Python + FastAPI + LangGraph backend, Gemini agents on Vertex AI, backend on Cloud Run. Honest note: it's a research-style hackathon project — the levels look charming, but game feel is still generator-grade.

V@VIBEFIX
0
0
2
0
mazkir
Showcase
Online
Developer Toolsclaude-code

mazkir

One shared memory for all your AIs, built mostly by Claude Code: tell one agent once and ChatGPT, Claude and Claude Code all remember.

Every AI you talk to starts from zero. mazkir is a shared memory layer that fixes that: tell one AI something once, and it shows up in your next chat — in Claude, Claude Code, and ChatGPT. It's a remote MCP server; your memory is plain Markdown notes you can read, edit, export, and delete, with per-project permissions. The feature list is practical rather than flashy: 20 MCP tools, OAuth login, a 30-day trash bin in case an agent deletes something, dark mode. Free during early access. The author's showcase post is disarmingly honest about how it was made: "claude code built most of it for me." Launch week brought ~160 visitors and exactly one sign-up — the author's friend. The numbers are real; so is the product.

V@VIBEFIX
0
0
7
0
My Claude
Showcase
Online
Developer Toolsclaude-code

My Claude

A private local dashboard for everything Claude Code keeps on your machine — skills, token spend, session history and costs, all in one readable place.

Every Claude Code session leaves traces on your machine — skills you installed, MCP connectors you wired up, tokens you burned, conversations you had. My Claude puts all of it on one private dashboard that runs entirely on your local machine. One command installs it: the script sets up an isolated Python environment and opens the dashboard on localhost. The Overview page shows your skills, conversations, 30-day token usage with API-price cost estimates, and disk use at a glance. Drill into Usage for live plan limits, Skills to see which ones you never touch (and push them to GitHub with one click), MCP & Plugins with a one-click health check, and History with full-text search across every session — including transcripts Claude Code has already rotated away. Nothing leaves your machine: it binds to 127.0.0.1, computes everything from files already on your disk, and secrets in MCP configs are auto-redacted before they're shown. The README ends with three words that explain how it was built: "Built with Claude Code."

V@VIBEFIX
0
0
7
0
Reachy Mini Vibe Coding Apps
Showcase
Online
Developer Tools

Reachy Mini Vibe Coding Apps

A conversational agent that helps you vibe-code tiny browser apps for your Reachy Mini robot

Reachy Mini Vibe Coding Apps is a conversational agent: describe what you want, and it helps you vibe-code tiny static browser apps for Reachy Mini robots, then publishes them as Hugging Face Spaces in one click. Updated 2026-10-10, running on Hugging Face Spaces. From the author's README: "A conversational agent that helps you vibe-code tiny static Reachy Mini apps for Reachy Mini robots, then publishes them as Hugging Face Spaces in one click." Built on the @pollen-robotics/reachy-mini-sdk as a 3-file static Space (author-stated).

V@VIBEFIX
0
0
9
0
BootLoops
Showcase
Online
Developer Toolsclaudeclaude-fable-5

BootLoops

Open-source LLM harness for exact quantitative science across 22 fields — code written by Claude, supervised by a Harvard physicist.

BootLoops is an open-source (MIT) harness that helps large language models do exact, interdisciplinary quantitative science: 49 tool packages spanning 22 quantitative science domains, written in Python 3.12 and Julia. In about three months the project produced 36 manuscripts and solved 15 previously unsolved elliptic Feynman integrals. The vibe-coding credential is in the repository LICENSE itself: "Created by Matthew D. Schwartz; code written by Claude (Anthropic) under his supervision." Schwartz, a Harvard physicist, wrote up the project as a guest post on the Anthropic Science Blog (October 1, 2026), describing how his experiments doing research computation with Claude crystallized into this software and its protocols. Version 1.0 shipped October 1, 2026 and is actively maintained. 中文导读：哈佛物理学家 Matthew Schwartz 发布的老牌开源 LLM 科学计算 harness——22 个定量科学领域精确计算，三个月产出 36 篇手稿；仓库 LICENSE 明言代码由 Claude 在作者监督下编写。

V@VIBEFIX
0
0
9
0
piermux
Showcase
Online
Developer Toolsclaude-code

piermux

A Tauri GUI for managing tmux sessions across multiple SSH hosts, with a cross-host tree view and line-buffered typing.

piermux is a desktop (and Android) GUI for managing tmux sessions across multiple SSH hosts: a cross-host session tree on the left, an attached tmux session on the right, and line-buffered typing explicitly designed for feeding prompts to AI agents. MIT-licensed, 25 releases deep since its 2026-04-28 debut. The README says it outright: "This is a vibe-coded side project — 'small things decide yourself, big things ask first'. Built with Claude Code as collaborator," with CLAUDE.md and NOTES.md keeping the whole collaboration public. Declared stack: Tauri 2 (Rust + WebView), russh, xterm.js, React 19, TanStack Query, Tailwind 4, SQLite, keyring 3. 中文导读：跨多台 SSH 主机管理 tmux 会话的图形工具，为 AI Agent 提示词输入优化的行缓冲打字。

V@VIBEFIX
0
0
6
0
boxes.dev
Showcase
Online
Developer Toolscodexclaude-code

boxes.dev

Cloud computers for Claude Code and Codex: give every agent its own machine, start it from Slack.

boxes.dev gives every AI agent its own cloud computer: bring your Codex or Claude subscription, start agents from Slack and Linear, and manage them from your desktop, phone, or terminal. The pitch is one line — "Close your laptop again." The makers, Nick & Drew, built it the way they dogfood it. Their Product Hunt maker comment (2026-10-06 launch): "We spent the last year coding almost exclusively with Codex / Claude Code" and "we've been building boxes.dev with boxes.dev for months" — the product that hosts coding agents was itself coded by coding agents, running on its own infrastructure. A neat meta-loop for the vibe-coding era: the tooling that lets agents work while you close your laptop was made by people who barely opened theirs. curated by the VibeFix editors

V@VIBEFIX
0
0
12
0
minecraft-modding-mcp
Showcase
Online
Developer Tools

minecraft-modding-mcp

An MCP server for AI-assisted Minecraft modding: 41 tools, 9 resources

minecraft-modding-mcp is an MCP server for AI-assisted Minecraft modding: 41 tools and 9 resources that let a coding agent inspect decompiled source, resolve Mojang/Yarn/Intermediary mappings, diff versions, analyze Fabric/Forge/NeoForge mod JARs, and validate Mixin and Access Widener files. Published on npm as @adhisang/minecraft-modding-mcp. The README states it plainly: "Note: This project is entirely vibe-coded — built with AI-assisted development without formal specs." 8 stars, 520 commits, last push 2026-09-25, MIT — a meta-vibe project: AI-built tooling that helps AI build Minecraft mods.

V@VIBEFIX
0
0
9
0
OtoDock
Showcase
Online
Developer Toolsclaude-code

OtoDock

Self-hosted 'company OS' — multi-tenant AI agent platform running Claude Code / Codex on your own servers

OtoDock 是自托管的"公司操作系统"：多租户 AI Agent 平台，让 Claude Code / Codex 以部门制协作运行在自己的服务器上，支持电话拨打、文档编辑、定时任务和沙箱隔离。 作者 Dimitris 在 Show HN 首发帖写了个罕见的声明："AI Disclaimer: large parts of OtoDock are written using OtoDock itself, running Claude Code." 用自己做的 Agent 平台写自己，递归式 dogfooding。GitHub 仓库 190 star、15 releases，最近 commit 是 2026-10-07（v1.7.1），技术栈 Python (FastAPI)、React + TypeScript、Docker、PostgreSQL，外加 bubblewrap + pasta 沙箱隔离。

V@VIBEFIX
0
0
10
0
pwasm
Showcase
Online
Developer Toolsclaude-opus-55

pwasm

纯 Python 实现的 WebAssembly 引擎：零依赖，安全沙箱跑不可信代码

pwasm 是 Simon Willison 的纯 Python WebAssembly 引擎：零外部依赖，可在沙箱里运行不可信代码，内置 MicroPython / QuickJS / Micro QuickJS 访客解释器，顺手带了 WASI 支持。作者自己管这叫 folly project——玩票项目。 作者在 2026-10-01 的博客原文写道："pwasm is one of my folly projects - an entirely vibe-coded pure Python WebAssembly engine that I built in January during my first bout of AI mania." 一月份 AI 狂热期起了个头，9 月底又把 Claude Opus 5.5 放进来冲刺了一轮（两天 42 个 commit），直接打到 Release 0.2a0，"几乎覆盖全部 WASM 规范"。仓库 10 star、Apache-2.0。

V@VIBEFIX
0
0
13
0
getrequest
Showcase
Online
Developer Toolsclaude-code

getrequest

API 可靠性层：流量中继、请求留存、失败一键重放，再不丢任何 API 请求

getrequest 是挡在后端前面的 API 可靠性层：流量中继、全量请求留存、失败请求一键重放。首页只写了一句话——"Never lose an API request." 流量高峰、宕机、失败的发布都不再意味着请求丢失。 作者 santoshbaggam 在 2026-10-09 的 Show HN 首发帖标题里写明："I claude-coded in a few weeks what I wanted to build since 2023"——从 2023 年就想做的东西，用 Claude Code 几周做成了第一版，正在找处理过这类问题的开发者试错。免费版永久免费、无需信用卡。

V@VIBEFIX
0
0
10
0
AI Animation Skills
Showcase
Online
Developer Toolsqclawclaude-code+1

AI Animation Skills

12 个 AI Agent 动效 Skill：一句话生成 PPT 翻页、流程图、架构图、视频分镜等炫酷 HTML 动画

AI Animation Skills 是一套「让 AI 做动画」的 Agent Skills：装好之后，对 AI 说一句话，就能生成带炫酷动效的完整 HTML 页面。12 个 Skill 覆盖 PPT 翻页、流程图、网络协议可视化、动态架构图、学霸笔记、卡片剧场、手机 UI 演示、视频分镜、Win11 仿真、瑞士风图表等场景——需求到成品之间，只剩一句话的距离。 仓库 README 明示，这是「基于 WorkBuddy / Claude Code / Cursor 等 AI Agent 的动效 Skill 集合」；其中的 dynamic-archify 更是作者用 AI 编程 Agent QClaw 魔改开源项目 archify 而来：静态图表变成带流动粒子动画、明/暗双主题切换、一键导出 PNG/GIF/WebM 的动态架构图。项目 4 月创建、10 月 5 日仍在加新 Skill，已收获 1466 star。对 AI 视频与科普创作者而言，这就是一条现成的动画生产线。

V@VIBEFIX
0
0
15
0
LDRAW Nova
Showcase
Online
Developer Tools

LDRAW Nova

Agent tooling that generates buildable LEGO models — vibe-coded with GPT-6 Astra and Claude Opus 5.5

LDRAW Nova is agent tooling for generative LEGO model building — and its author says the idea only became buildable when the models caught up: generating LEGO models that are actually valid "proved really hard to get right, even if vibe coding it... until GPT-6 Astra and Claude Opus 5.5 arrived," at which point the project itself was vibe-coded. The output is Python code that produces mathematically valid models, exportable as Blender-editable glTF — the showcase render is a Sakura Garden pagoda scene assembled from real parts. The repo includes agent instructions, model-checking scripts and worked examples, treating the AI agent as a first-class collaborator rather than a code autocomplete. It's a neat inversion of the usual story: here the vibe-coded artifact isn't an app but a toolchain that lets agents do physical-world-valid design. 309 stars and commits this month suggest the LEGO-builder niche is very much alive.

V@VIBEFIX
0
0
14
0
CoIsland
Showcase
Online
Developer Toolsclaude-code

CoIsland

Your whole engineering stack, centralized in the Mac notch — 35+ connectors, one click hands any alert to your coding agent.

CoIsland lives in your MacBook notch and watches your engineering stack: 35+ connectors for GitHub, Vercel, Snowflake, Jira, Linear, Sentry, PagerDuty, email and calendars. Each monitor runs on its own schedule, and when something fires — a failed deploy, a red CI run, a failed payment — the notch shows the alert with what was found. One click hands it to Claude Code, Codex or Cortex Code, which opens in a terminal right under the alert with the full context; the agent does what is safe, asks you for the rest, and never closes an alert — you do. It is a native Swift/SwiftUI app for Apple Silicon, deliberately serverless: tokens live in the macOS Keychain and the app talks to your tools directly. Two monitors run free forever; a license is a one-time $29.99 early-bird purchase. The maker has run his own days from the notch since July 2026. The author's own disclosure: CoIsland was built with Claude Code (Opus 5.5). curated by the VibeFix editors

V@VIBEFIX
0
0
13
0
proxy429
Showcase
Online
Developer Tools

proxy429

本地 LLM API 代理：自动吃掉 429 限流和抖动，让 Claude Code / Codex 的长会话不再被上游打断。

proxy429 是一个跑在本地的 LLM API 代理，夹在 Claude Code、Codex CLI 等 AI 编程 agent 和上游模型提供商之间，专门解决一个真实痛点：跑第三方兼容端点时频繁的 429 限流和流式中断。它对 429/5xx 做指数退避自动重试（含藏在 HTTP-200 SSE 流里的限流错误），SSE 保活让客户端在代理重试期间永不超时，还支持模型路由与劫持、OpenAI Responses 与 Anthropic Messages 双协议翻译、加密搜索信封（后续提问直接读上次搜索内容）、以及把显式关思考的请求悄悄升级为 low 思考的细节优化。 配套的本地网页控制台是亮点：实时查看在途流、token 与缓存命中统计、逐请求诊断，请求/响应体按“代理到上游”和“客户端到代理”两侧分开记录——排查 agent 到底发了什么、收到了什么，一目了然。控制台中英双语，跨平台托盘应用单二进制，约 40% 测试代码，MIT 开源。 作者 aaahzb 的原话：这是一个“为 AI agent 造的工具，由 AI agent 造”——他不是 Go 开发者，整个项目是纯 vibe coding 产物。 curated by the VibeFix editors

V@VIBEFIX
0
0
13
0
BagIdea Office
Showcase
Online
Developer Toolsclaude-code

BagIdea Office

Your desktop wallpaper becomes a living AI-agent office — real Claude Code sessions that walk, work, delegate and hold meetings.

BagIdea Office turns your desktop wallpaper into a living HD-2D office where every pixel-art character is a real Claude Code session. You play CEO: give an order, a Director agent breaks it down and delegates, and you watch the actual work happen — characters walk to their desks, open files, and run tools in real project folders. A glowing monitor means a live tool call is running; a character walking over to Security means a real permission request. Nothing on screen is canned animation. The office extends itself: agents propose plugins and skills (you approve each one), accumulate shared knowledge in OFFICE.md, and even hold meetings on their own that can turn into written project proposals. It talks over voice, Telegram, Discord and LINE, speaks 14 languages, and lets each agent run a different model brain — Claude, GLM, DeepSeek, Qwen, OpenAI, Gemini, or local Ollama. The project is heavily maintained (241 stars, 98 releases) and MIT licensed. The author's own disclosure: the app was itself built with Claude Code. curated by the VibeFix editors

V@VIBEFIX
0
0
9
0
termist
Showcase
Online
Developer Tools

termist

Mission control for your coding agents: Claude Code, Codex and OpenCode side by side in one terminal.

termist is a terminal mission control for people who run multiple coding agents at once. Every session becomes a card with a dot: running, waiting for you (a permission or a question), done-unseen, done-seen. Jump straight to the agent that needs an answer, send follow-ups without opening its terminal, resume sessions by name, manage projects as tabs — and the agents keep running when you close the window, owned by a small daemon. The author's disclosure: AI coding tools were used throughout the project — implementation decisions, testing and project direction are the author's. It is at v0.1.4 and moving fast: GitHub PR review (comments, line threads, reviews) already landed, installable on macOS, Linux and Windows. Plus the best-looking terminal UI of the year: Istanbul scenes (Galata Tower at dusk, the Bosphorus) that change with the hour. curated by the VibeFix editors

V@VIBEFIX
0
0
8
0
MechFaber
Showcase
Online
Developer Toolsclaude-code

MechFaber

Describe a machine in words — AI picks the parts, writes the CAD and firmware, then simulates it all together.

MechFaber is an AI machine-design engine: you describe the machine you want in plain language, and it searches vendor catalogues for motors, bearings and controllers you can actually order today, draws the CAD around the suppliers' own 3D models, validates fit and load, lays out the full electronics loom, writes the firmware, and then runs the entire machine in simulation — real firmware binary on an emulated MCU (Renode) coupled tick-by-tick with physics (MuJoCo), sharing one battery model. When something is wrong, you reply in a sentence and the whole loop re-runs. The author's own description of the build: Claude Code did all the design work including the firmware. Built over roughly 3 months with a split-agent setup — separate agents for research, calculation, CAD, electronics, firmware and simulation, after a single do-everything agent failed. Clone machines from the library and rebuild them locally, or publish your own. Free while it is early. curated by the VibeFix editors

V@VIBEFIX
0
0
9
0
GHOSTVAULT
Showcase
Online
Developer Toolsboltnew

GHOSTVAULT

Enterprise-grade dark-web breach monitoring with zero-knowledge architecture — built in 72 hours on Bolt.new.

GHOSTVAULT is a breach-monitoring platform that aggregates 5+ leak data sources (HaveIBeenPwned, DeHashed, IntelligenceX and more), encrypts everything browser-side with AES-256-GCM so even the operator can't see your data, and fires real-time dark-web alerts. The author, Alex Sofroniev, built it to prove 'enterprise-grade cybersecurity could be built rapidly using modern no-code platforms' — 72 hours, $0 development cost, and the project grew so technically complex it started hitting Bolt.new's own platform limits. Pricing runs from €6.90/month Pro to $299/month Enterprise, and it entered through Bolt's World's Largest Hackathon. curated by the VibeFix editors

V@VIBEFIX
0
0
9
0