New releases

Show HN: Pace or Pause – Explore where people stand on AI development

Show HN: Tenbri, a log AI agents can’t secretly rewrite

Show HN: Modula-9, a Wirth-family language designed for auditing AI-written code

Show HN: Not Quite My Tempo – A Whiplash-Style AI Reviewer for GitHub PRs

Show HN: Koody for Business – AI bookkeeping for people who aren't accountants

Show HN: Worklicant – Lightweight ATS. Pay per Vacancy. AI Enabled

Show HN: LongTerMemory, AI-RAG Study tool helping people to study everything

Show HN: Agent Alley – A social experiment for AI agents

Show HN: Standard Topic – An AI and RSS-powered news and resource platform

Show HN: Reverse Captcha, prove your agent is AI

Show HN: Doozy – an AI coworker in Slack that works in prod without your keys

Show HN: Tori by TalkJS. AI support that's actually good

Show HN: Shazam but for AI Slop Music

Show HN: MarkItDown AI – Convert documents to Markdown locally in the browser

Show HN: Sandkiln – Fast Firecracker MicroVM sandboxes for AI-generated code

Show HN: I built an AI tool that turns two photos into a dance video

Show HN: AIExportHub – bulk export AI chats to Notion, Markdown, and PDF

Show HN: A practical framework for evaluating agentic AI systems

Show HN: Visual heat map of AI news updated hourly

Show HN: Search for AI-agents and usefulness score

Show HN: AppMaker – Build business apps with AI, visual editing and code

Show HN: Self-hosted AI spend attribution

Show HN: Chai-UI – A design system to generative AI apps

Show HN: What the Harness? LLM agent inconsistencies in MCP implementation

Show HN: Skillsync – Coding agent memory but it's just search

Show HN: My attempt to get Claude to run a business

Show HN: BodyCommit – movement breaks in VS Code while your agent works

Show HN: Seclai – See what your LLM agents are doing in production

Show HN: Faztudo, a local chat for coordinating the coding agents CLIs

Show HN: AML CYCLE 2, a benchmark for long-term agent memory

Show HN: barn.app – a digital stream deck like agent monitoring tool for macOS

Show HN: Craig – A TUI for running parallel coding agents

Show HN: Jev in action – task classifier and auto model router in Keen Code

Show HN: A better way to use DNS-AID to publish your I agent

Show HN: Buddylist – Coding agents in an MSN-style messenger

Show HN: Acyclic; Rapid agent swarms

Show HN: A daily game where the goal is to be unpredictable to an LLM

Show HN: Rocus: Browser local LLM that organizes your websites, notes and files

Show HN: SentinelGate – Drop prompt injections at Token 0 before LLM inference

Show HN: A free MCP server for facts LLMs get wrong (time zones, holidays, FX)

Show HN: MeetCap – Free on-device meeting transcription your AI agent can run

Show HN: Independent LLM benchmark index tracking 22 frontier models

Show HN: Benchmarking LLM ability to code in an adversarial language

Show HN: Risphai – picks the AI model, tools and prompt for each request

Show HN: Sys2txt – transcribe system audio to text in the CLI with a local LLM

Show HN: Llmspeed – feel how fast a local LLM runs before you buy the hardware

Show HN: MCP Census – a live change feed for 48,797 MCP servers

Show HN: A better hosted MCP server for QBO and Xero

Show HN: MCP for Ordering Assembled PCBs

Show HN: TaskPocket, a to-do app with an MCP server instead of built-in AI

Show HN: Spomory – graph-based MCP memory with an RL-trained forgetting policy

Show HN: MCPX – MCP server directory scanning for tool poisoning, leaked keys

Show HN: I benchmarked 6 MCP proxies in Claude Code, then built my own

Show HN: I built 7 B2B APIs and 5 MCP servers for compliance and trade

Git designed for AI agents.

Show an agent a problem: mark up, record, send

Code graph and memory for your coding agents, in one Rust program. For Claude Code, Codex, Cursor and any MCP client.

Agent skill to reverse engineer a SaaS product through the browser into a knowledge base for rebuilding it: screens, rules, data model, design tokens and production barriers, without touching production data.

Awesome Reverse Engineering Anything: tools, projects, case studies and research papers for websites, games, apps, binaries, firmware and communication protocols.

Accessibility-first browser & computer use for AI agents — one MCP server for every harness (Codex, Claude Code, OpenCode, Antigravity). Instruments the DOM/accessibility tree via CDP instead of screenshot loops: diffed snapshots, stable element refs, trusted input, typed errors. Zero dependencies

touri: open-source desktop app for AI agent bots (Codex, Claude, OpenCode) with isolated VMs, browser and computer use.

Open GRC skills for AI agents. By ISMS Copilot.

AGENTS.md contract for AI coding agents + zero-dependency linter + SessionStart hook. Works with Claude Code, Codex, Cursor, Copilot.

⚡ The fastest open-source AI gateway — one Rust binary with a built-in console to route, secure, budget and observe every LLM call across OpenAI, Anthropic, Gemini, Bedrock and more.

A language model trained from scratch and running offline on an ESP32-S3 with a touchscreen chat interface.

全网最全的 AI 破甲合集：一个仓库破甲所有模型。收录 1733 个上游破甲项目，9 条可一键安装的破甲卡覆盖 GPT/Claude/Gemini/DeepSeek/GLM/Kimi/Grok/Qwen 与本地模型，内置 llms.txt 让 Claude Code/Codex/Cursor 自己装 有用的话可以点个收藏星星，会持续收录

Skill for AI agents that draws publication-quality scientific schematic figures as vector code

Reason Network: your team's coding agents, talking to each other. The engine behind `reason network`.

Local AI BGM and sound effects for Claude Code and any MCP client. Stable Audio 3 on Apple Silicon (MLX) and Windows NVIDIA GPUs (PyTorch). No cloud, no API key.

Analyze the Indonesian stock market (IDX/IHSG) from web terminal, CLI, or any AI agents via MCP server. IHSG Prices, Fundamentals, and Screening. Free and No API keys required. Changelog: https://github.com/fzn0x/idx/blob/main/CHANGELOG.md

你今天还能吃多少卡，取决于你愿意走多远 —— 用步行距离换卡路里额度的吃饭决策。麦当劳官方营养值 × 周边餐厅，算你这顿还吃得下什么。

Self-hosted AI workspace for software engineering capstone and course projects. Generate UML models, system designs, test cases and editable DOC specifications from requirements. Supports traceability, your own model provider and optional read-only MCP access.

麦麦现拼：到店自取、同店陌生人现场拼一个大套餐（麦当劳 MCP 开发者大赛）

麦门语文课 · 接入麦当劳官方 MCP（只读 order-list 读取真实订单），把餐品写成一篇语文课文。16 篇原创文学戏仿，纯 Python 标准库实现、不调用任何大模型；另附零依赖的网页版可离线体验，并一键导出手机分享图片。麦当劳程序员节创意开发大赛参赛作品。

麦有谱：同样的麦当劳，不同的吃法。基于麦当劳 MCP 的组合吃法 Skill 与交互演示。

Debloat any Android or Google TV over Wi-Fi and get an Apple TV-style home.

Terraria 1.4.5.8 Web Rewrite, using GLM-5.3-Flash and GLM-5.3

Build a software startup without writing code. Not vibe coding: a proven team workflow (requirements, design, test-first build, QA, release approval) run by AI agents.

Multiplayer mode for AI coding agents. Your team's Claude Code, Codex and Cursor agents huddle up: live status, one master prompt, a shared task board, and no stepping on each other's files.

kulkas2pintu/wan777

hugging-apps/bfs-best-face-swap

black-forest-labs/flux-3-image-klein

multimodalart/jev-decision-index

OpenMOSS-Team/MOSS-Image

fishaudio/Drama-3

AST-1320/Qwen-2511-Plus-Flux.2-klein

pablodawson/minimax-h3-360-orbit

wank3r/minimax-h3

hugging-apps/qwen-qwen-image-2-1-turbo

glitchlab/MiniMax-H3-Turbo-Lora-UNCENSORED-Workflow

raubatz/qwen-image-2.1-uncensored-aio-loras

Aura1in/Krea-2-Turbo_v2

hitman911/wan2-2-fp8da-aoti-preview-2c-finishing

Viggle/Qwen-Image-2.1-viggle-turbo

vamo455/Omni-videos-custom-auto_prompt_high-quality

ibm-granite/speech-to-decisions

fishpasters/uncensored-imageTOimage2Ref

tardellirs/model-pulse

lightonai/LightOnOCR-3-Demo

selfit-camera/Omni-Image-Editor

Qwen/Qwen-Image-2.1

microsoft/TRELLIS.2

hugging-apps/minimax-h3-character-swap-lora

Lynote/free-ai-humanizer

Edisoneh/InfiniSplat2

aet256/Qwen-Image-Edit-Rapid-AIO-Loras-Experimental

kulkas2pintu/QWEN_EDIT_IMAGE

JCisHERE/wan2-2-i2v-v3

speridlabs/iris-3b

deepseek-ai/DeepSeek-V4.1-Flash-1017

black-forest-labs/flux-3-image-klein-comfy

fishaudio/drama-3

meta-models/Muse-Spark-Open-816B-A42B

google/embeddinggemma-2

black-forest-labs/flux-3-image-klein

OpenMOSS-Team/MOSS-Image-SFT

jialinyyzz/humanizer

Lightricks/LTX-2.5

OpenMOSS-Team/MOSS-Image-Base

abenzerps/Qwen-Image-2.1-Uncensored-Quantized

LiquidAI/LFM2.5-24B-A2B

Qwen/Qwen3.8-27B

Comfy-Org/MOSS-Image

nvidia/Qwen3.8-Flash-Next-FP4-Local

datalab-to/indra-ocr

tencent/Hy-Memory-30B-A3B

mistralai/Timestral-209M

facebook/sam3

Qwen/Qwen-Image-2.1-Turbo

Harness the agent. Then harness its self-improvement.

Fast and affordable AI video for high-frequency creators

The video studio where AI agents edit with you

The AI knowledge graph that builds itself

The coding agent that runs on its own AI models

Your emotional wellness AI. A friend to figure it out with.

Build and edit no-code apps inside ChatGPT & Claude

Write a to-do and your AI agent gets it done

Smarter search that gives AI agent richer data at lower cost

Simulate an AI agent community reacting to "what if"

Every friends' chat gets an AI friend with superpowers

A modular development board built for agents

Open models, built for agents.

The easiest way for AI agents to pay for their own tools

An always-on agent that keeps your docs up to date

Give your AI agent a phone that calls and answers

AI agents on their break, performing a live talk show

Memory-rich AI inference without HBM

Point at your screen and start yapping to your agent via MCP

Know when to push your coding agents, and when to ease off

Talk to your AI coding agent over a walkie-talkie

Run your AI coding agents on an infinite canvas.

Give your AI agent 31 web tools in one MCP server

Highly efficient agentic 501B parameter open model

Run AI agents on one Mac and use its windows on another