Explore AI Tools & Apps

Pie

Pie Labs · Coding

Pie is an AI testing tool that walks through your web and mobile apps the way a real customer would, then flags the bugs it finds. It runs end-to-end testing without asking you to write scripts. Instead, it maps your user flows on its own and reproduces them every time you push code. When something breaks, Pie opens a pull request with a suggested fix so your team reviews a change instead of a bug report.

Teable

Teable · Coding · Productivity

Teable is an AI no-code database that turns plain-language prompts into working tables, apps and automations. It pairs a spreadsheet-style interface with a real PostgreSQL engine, so non-technical teams can build business apps while developers keep a database they can query directly. It suits teams that have outgrown scattered spreadsheets but don't want to hire a full engineering team. Think of it as an AI spreadsheet that behaves like a database.

Vicoa: Claude Code & Codex IDE

Vicoa · Coding

Vicoa is a remote control for AI coding agents. The product, also called Vicoa Code Anywhere, runs Claude Code, Codex, OpenCode, and a dozen other CLI agents on your own Mac, Windows, Linux machine, or server, then lets you start, watch, and steer those sessions from a phone, a tablet, or the web. Your code never migrates to Vicoa's servers. What moves is the conversation. Prompts, diffs, and approvals sync across every device you're signed in on. Think of it as a mobile AI coding assistant that keeps your desktop agents within reach. It's built for developers who kick off long agent runs and don't want to babysit a terminal all afternoon.

Airbolt AI

Airbolt AI · Coding

Airbolt AI is a developer service for LLM integration without the server. You drop its SDK into a React or JavaScript/TypeScript client, point it at your own provider API keys, and start calling OpenAI, Anthropic, Google Gemini, and other large language models in minutes. That's a zero-backend AI SDK in practice: no server to run, no keys in the browser. The company handles tokens and rate limits for you. It sells itself as the fastest path to a working AI feature.

Google Chrome with AI

Google · Coding

Google Chrome with AI is Google's flagship browser rebuilt around Gemini, its multimodal AI model. Instead of a separate chatbot window, the assistant sits inside Chrome as a side panel and reads the pages, tabs, and history you already have open. It can summarize a long article, compare the specs on three shopping tabs, rewrite text you select, or run multi-step tasks across sites through a feature called Auto Browse. The pitch is simple. Your browser already holds the context of what you're doing, so putting the AI there removes the copy-and-paste step that every standalone chatbot demands. Google ships the AI browser to Windows, macOS, Android, and iOS, with the deepest feature set tied to a paid Google AI subscription. Want the AI to see your tabs? That's the whole idea.

Tyran AI

Terra · Coding · AI Assistant

Tyran AI is an AI health agent builder for apps. It connects to more than 500 health metrics from wearables, labs, nutrition, sleep, and HRV data, then turns that stream into features your users can act on. Instead of wiring up raw sensor feeds yourself, you pick an agent template or write a prompt, define a schedule and an outcome, and get typed JSON back. The pitch is simple. Real health insights in minutes, not months. It's a health metrics API with an agent layer on top.

Sudo AI

Sudo · Coding · Other

Sudo AI is a unified API that lets developers route requests across multiple large language models, including OpenAI, Anthropic, and Google, through a single endpoint. Instead of wiring up each provider separately, you send one call and Sudo handles model access, fast routing, and usage metering. It also adds context management and monetization tools like usage-based billing and in-app ads. The pitch is straightforward. A small team can launch and charge for an AI product without building billing from scratch.

stackoverflow.ai

Stack Overflow · Coding

stackoverflow.ai is Stack Overflow's AI-powered developer search and discovery tool. It blends generative AI with the site's 16 years of human-verified answers, so you can ask a technical question in plain language and get a grounded response that points back to community knowledge. It runs in the browser, needs no install, and is free to use.

MCP Snitch

Adversis LLC · Coding

MCP Snitch is a security monitor for MCP tools that watches every call an AI agent makes and decides, in real time, whether to allow it. Built by Adversis, it protects API keys, hardens MCP server security, and keeps a full audit trail you can use to audit MCP tool calls later. It runs on macOS, plugs into VS Code and Cursor, and keeps all analysis on your machine.

Cardinal

Cardinal · Coding · Productivity

Cardinal is an AI revenue platform that lets high-growth companies build GTM agents for outreach, follow-up, research and scheduling, all from one workspace. Agents also handle CRM syncs. Instead of juggling a list builder, an email tool and a sequence app, teams define a target list, launch agents that run a specific motion, then keep whatever pulls in meetings. It suits founder-led sales and lean revenue teams that want outbound automation without stitching seven tools together.

Capalyze

Capalyze · Coding

Capalyze is an AI data analytics tool that doubles as a web scraping tool. It turns messy web pages into structured tables, then lets you analyze them with plain-language questions. It runs as a Chrome or Edge extension, so you can pull product listings, reviews, posts, or price pages straight from your browser without leaving the tab you're already on. Ask it to group, chart, or summarize what it found and you're done. If you've ever copied comments into a spreadsheet and then lost an afternoon cleaning them, this is the shortcut.

Model Kombat by HackerRank

HackerRank · Coding

Model Kombat by HackerRank is a free AI coding arena where you watch anonymized coding models fight it out on real programming problems. You pick a challenge, two models answer it, and you decide which response is better without knowing who built either one. It's built for developers who want a straight answer to a simple question: which coding model actually helps me write working code? The platform comes from HackerRank, a company that's been running coding assessments and skill tests for over a decade, so the setup leans on real programming tasks rather than recycled puzzle sets.