AI Agents1 month ago
Deck | Design and Deploy Computer Use Agents
Design and deploy AI agents that automate tasks across software without requiring APIs.
@gabebenchimol · X
The full gallery
Tech stack
60 projects
Design and deploy AI agents that automate tasks across software without requiring APIs.
@gabebenchimol · X
Try free REST APIs for QR codes, DNS lookups, and more in an interactive playground.
@SimpleWebAPIs · X
Add data persistence and authentication to AI-built apps in minutes without database setup.
@MuratOnX · X
Build production-ready web apps with AI workflows and databases without writing code.
@sanchoyai · X
2/ Meet Momen - the no-code builder for production-ready Apps Go to → Got a big app idea but can’t code? The no-code builder that lets non-technical founders create apps go from idea → millions of users. UI, database, workflow, AI — all in one. Not vibe coding, but something you can control, grow, and scale.
Create and run always-on Hermes agents without managing servers or infrastructure.
@RyanRayMartin · X
Create custom apps and tools in your browser without writing code.
thejarren · HN
Hey HN, I'm really excited to share Emra with you. The yc description would be: It's like "Notion meets Lovable". It's a workspace where you build full apps using AI and every app sits on one shared database and service layer. Generated apps are always editable, and (coming soon:) can interact with each other in the workspace. I'm of the opinion that the cost of software will go rapidly to zero (I don't think that's too surprising to any of you), but there will always be room for effort and creativity. Emra is essentially built for the future where every person is a developer. Essentially personal software for everyone. So far I've built notes apps with a 3d graph view to see connections, a video editing app and a canvas design tool (figma clone) with export capabilities, as well as a collection of project management tools and games. Right now the most satisfying thing is being able to fix a bug I run into in around the same amount of time that it takes to send a bug report in any tr
Single API key for accessing OpenAI, Claude, Gemini and other AI models.
@lologfts · X
built Apiarium, with one api key you can choose multiple Ai models, easy integration
Convert OpenAPI specs and databases into MCP tools without coding.
mhmt_dmr · HN
MCP Gateway – Turn existing APIs and databases into MCP servers
Publish private, public, or shareable iPhone widgets powered by your content, automations, and APIs. No app required.
@talkravarusic · X
Building Glance, it’s lets agents step out of the chat box and into your iPhone Home Screen
Backend platform providing auth, storage, database, and LLM APIs (Claude, OpenAI, Gemini) via a single JavaScript library.
ent101 · HN
The Safe, Production-Ready Backend for AI-Generated Apps
Lower your AI costs through one developer-friendly API platform. Start with ChatGPT and Claude, transparent usage billing, and familiar protocols.
@kiraroute · X
Quick Start is now built into KiraRoute. ⚡ Set up Node.js, Codex, Claude Code, OpenClaw, or Hermes—or copy ready-to-run API examples. Dynamic endpoints, one-click copy, mobile-friendly, in English and Chinese. →
Access open AI models via OpenAI-compatible API using crypto payments without account creation.
@vagaziner · X
AI access still starts with a form. email card KYC an account you will forget a 5% tax sitting on top of the model bill @iamzhe built the opposite of that stack. is one OpenAI-shaped endpoint. Paste the base URL into Claude Code, Cursor, Codex, OpenCode, whatever you already use. Drop in a key. Pay with USDC, SOL, or whatever Solana token is sitting in the wallet. Fifty cents is enough to start. No signup. No passport photo. Credits do not expire. They also do not withdraw, on purpose, so the product never has to become a bank. Two hundred thirty nine models behind one key. Llama, DeepSeek, Qwen, Gemini, GLM, the weekly flavor of the month. If you do not want to pick, send minirouter/auto and let the gateway choose. If a provider dies mid-call, it fails over and tells you who actually answered and what it cost. OpenRouter takes five percent. Minirouter takes zero on the deposit and zero markup on open-weight routes. Ten dollars in is ten dollars of inference at