Explore AI Tools & Apps

box

ASCII · Coding

box is a cloud sandbox service from ASCII that gives AI agents their own persistent Ubuntu virtual machines. Think of it as an AI agent sandbox, not a shared container pool. Each sandbox runs as a full Linux VM with Docker, SSH and SCP access, a dedicated IPv4 or IPv6 address, and a 60fps desktop you can open in a browser. You spin them up with a CLI, API or SDK, and stop paying the moment a sandbox stops running.

Routebase

Routebase · Coding

Routebase is an API lifecycle management platform built around one living OpenAPI spec. Think API design tool, API documentation platform, and API mocking tool rolled into one contract. You design that contract, publish it, and Routebase generates the docs portal and mock server while your tests and monitors validate against the same version. Routebase keeps every downstream tool honest, so the thing you documented is always the thing you actually shipped to consumers. It ships an MCP server too, so AI agents design, lint, and test APIs through the same workflows your team already uses. Routebase handles the busywork, and you handle the design decisions.

Rerun

Rerun · Coding

Rerun is a no-code AI agent automation platform that lets you describe a job in plain words and have an agent set itself up to do it, connecting to tools like Gmail, Slack, and Notion and running on Claude or ChatGPT on a schedule or a trigger. You watch runs, logs, and metrics from a live dashboard. Teams that want AI workflow automation without touching a terminal are the ones it's built for.

Manifest

Omfang AB · Coding

Manifest is a web action manifest API that turns any web page into a structured JSON action map of what an AI agent can actually do there: every button, form, and input, each with the fields it needs and the prerequisites that must be met first. Built by Omfang AB, it sits between browser tools that only render a page and scrapers that only read text. You send a URL, and you get back a machine-readable action manifest your agent can reason about directly. Think of it as the agent action layer for browser automation for AI agents.

ditto.site

ion-design · Coding

ditto.site is an open-source, MIT-licensed website cloner that turns any public URL into a working, componentized Next.js or Vite React project. Instead of asking a language model to guess at your markup, it captures what the browser actually renders and emits a deterministic codebase, usually in around five minutes. You can self-host the whole pipeline, call the hosted REST API, or wire it into an agent through its MCP server, and it costs nothing either way. Hard to argue with that.

CartAI

CartAI · Coding

CartAI is a developer-first API that deploys a specialized agent to complete transactions on the web. You send it a target page and a task, and the agent navigates that page, fills the cart, and runs the checkout through a PCI-compliant payment stack. It handles e-commerce checkouts, subscription sign-ups, invoice payments, and order submissions wherever a transaction has to clear, whether that surface is a public store, a custom cart, or a login-gated vendor portal. Think of it as an AI agent for payments that also knows how to shop.

Lunen.ai

REDspace · Coding

Lunen.ai is an AI agent control plane for the enterprise. It lets anyone on your team describe an agent in plain language, then turns that description into a structured execution plan with named tools, scoped data, and a schedule IT can approve. The pitch is simple: the same flow that builds an agent also sets its permissions and logs its actions, so enterprise AI adoption survives a security review instead of dying in one.

Autoplot

Kaya Aykut · Image · Coding

Autoplot is a native, local-first macOS workspace for scientific plotting and data analysis. It handles tabular data from import to export: parsing messy files, creating reusable variables, building 2D and 3D figures, running distribution and power-law fits, and exporting print-ready vector PDFs. Unlike browser-based tools, it keeps your raw data on disk and runs the scientific Python stack offline.

HeimWall

HeimWall · Coding

HeimWall is a macOS menu-bar agent that watches what engineers type into AI coding tools like Cursor, Claude Code, Copilot, ChatGPT Desktop and Windsurf. It detects secrets, PII and confidential data on the laptop, masks them on the record before anything is stored, and keeps a tamper-evident audit log your team can review. Think of it as AI prompt security without surveillance. Managers see categories and counts, never the raw text.

BaseRT

Base Compute · Coding · Other

BaseRT is a from-scratch local LLM inference runtime built for Apple Silicon, tuned at the Metal kernel level to squeeze more speed out of Mac hardware than general-purpose engines like MLX or llama.cpp. It ships as a single `basert` LLM CLI that pulls models from HuggingFace, converts them to its own `.base` format, and lets you chat locally or serve an OpenAI-compatible API. Developers who want fast, private, on-device AI on a Mac get the most out of it, because there's no cloud dependency, no API keys, and no per-token cost.

AgentGrid

AgentGrid · Coding

AgentGrid is an infinite-canvas desktop workspace built for coding agent orchestration. It puts Claude Code, Codex, Cursor and OpenCode on one pannable surface so you can run several agents side by side, hand each one a focused job, and keep terminals, source control, browsers and notes next to the work. It's a multi-agent workspace you can actually see. It runs on macOS, Windows and Linux, and you bring your own AI subscriptions. That last part matters.

ZooData

ZooData · Coding

ZooData is an agent-native commerce data layer that turns messy ecommerce signals into clean, structured JSON. It pulls from live and 2+ years of historical Amazon data, then hands the results to AI agents over an API-first OpenAPI v2 interface. Building an agent that reasons about products, prices, or market trends? ZooData is the plumbing that feeds it reliable data instead of scraped junk. That distinction matters more than it sounds.