Agents

Personal agent with a Bun gateway that keeps memory, skills and run history, with approval-gated tools and web, CLI and mobile clients.

Browser workbench that coordinates Claude Code, Codex, Gemini, OpenCode and other CLI agents as a visible team of local processes.

Open-source, local-first workspace where humans and AI coding agents coordinate through channels, tasks and persistent memory.

Hub that launches Claude Code, Codex CLI, Gemini CLI and other agents with third-party APIs and reaches them from web, mobile and chat apps.

Multi-agent research workbench that searches arXiv, OpenAlex and Semantic Scholar and drafts evidence-grounded literature reviews.

Local desktop and web workbench that reviews a repo's pull requests with Claude or Codex in isolated worktrees and posts only approved comments.

Claude Code agent on a VPS you own, with a real Chromium browser and a Telegram bot to message it.

Self-hosted AI agent harness for authorized penetration tests and security labs, using your own model API keys.

Open-source deep research agent for multi-step web research, with models published on Hugging Face.

Coding agent that runs fully in the browser on WebGPU, with a local MiniCPM5-2B model and a virtual shell workspace.

In-browser coding agent that writes web app files into a virtual workspace with a live preview, using a fine-tuned 4B model.

AI app builder from Convex that generates full-stack web apps with a built-in database, auth and background workflows.

Open-source agentic software engineer that plans, researches the web and writes code from a high-level task

Mobile-first web UI for OpenCode that connects to running OpenCode instances for sessions, chat, file mentions and model selection

General-purpose biomedical AI agent that autonomously runs research tasks by combining LLM reasoning with code execution.

Open-source company intelligence agent that looks up live sources such as the web, Slack and Drive and keeps its own wiki and CRM.

Personal research assistant for scientific work, built on LangChain DeepAgents with an isolated Docker sandbox.

AWS's AI assistant for developers that chats, writes and tests code, upgrades Java and .NET apps, and helps operate AWS

Multi-agent LLM framework for web search that runs several queries at once, with React, Gradio or Streamlit front ends.

Research system that searches the web, writes an outline and drafts a Wikipedia-style article with citations; Co-STORM adds human collaboration.

Local-first personal AI assistant with a gateway control plane, Telegram and Discord connections, a WebChat UI and a BSC trading action.

AI agent platform for business support that uses A2A and MCP standards to connect agents, apps and tools.

Wix's vibe-coding platform that builds apps, websites and AI agents from plain-language descriptions, with backend and hosting built in

Desktop agent that boots a sandboxed computer for each task and completes work by clicking and typing in the UI.