Companies in the category 'AI Agents'

OpenClaw is an open-source personal AI agent platform that runs on the user's own machine and integrates with messaging apps such as WhatsApp, Telegram, Discord, and Slack. Originally created by Peter Steinberger in November 2025 under the name Clawd, the project became one of the fastest-growing open-source repositories in history, accumulating over 290,000 GitHub stars. In February 2026, Steinberger joined OpenAI and transitioned OpenClaw into an independent foundation to ensure it remains open and community-driven.

Yao is an open-source, single-binary app engine for building autonomous AI agents and web applications. Developed by Infinite Wisdom Software, it provides a full-stack runtime that combines data modeling, REST API generation, agent orchestration, and a UI framework in one executable. Yao supports multi-agent collaboration with event-driven, proactive, and self-scheduling capabilities, and integrates natively with MCP tools and GraphRAG for knowledge retrieval.

Orkes provides an enterprise platform for building highly reliable applications and AI agents, enabling workflow orchestration across various clouds, languages, and frameworks.

Builderz is an AI and blockchain development agency that designs, builds, and operates production-grade AI agents, multi-agent orchestration systems, and blockchain applications for startups and enterprises. Founded in 2022, the company also develops and maintains open-source tools including Mission Control, a self-hosted AI agent orchestration platform, and the Builderz Solana dApp Scaffold.

Mastra provides a low-code, TypeScript-based platform that automates the process of deploying and managing AI agents for enterprises. Their technology simplifies AI integration, streamlining workflows and reducing the time and expertise needed to operationalize AI solutions.

OpenWork is a desktop application that provides an open-source alternative to Claude Cowork, enabling teams to run AI agents and workflows using 50+ LLMs with bring-your-own-key support. The platform includes a cloud-hosted sandbox service called OpenWork Den, allowing teams to share and run agentic workflows across desktop, Slack, and Telegram.

Notte provides a full-stack web agents framework for rapidly building reliable web agents that act and scrape on any site. It offers solutions for browser automation, web scraping, and automated workflows, including CAPTCHA solving and virtual credit cards.

Golem Cloud is an agent-native durable computing platform built on WebAssembly that enables developers to build and deploy AI agents and distributed applications with guaranteed reliability. The platform provides transparent failure recovery, exactly-once external effects, and durable state persistence, eliminating the need for complex infrastructure management. Golem Cloud is developed by Ziverge Inc. and supports TypeScript and Rust for writing scalable, durable agentic systems.

21st.dev is a platform providing UI components and infrastructure for building AI-powered applications, including a large registry of open-source React components and an Agents SDK for deploying AI coding agents with built-in sandboxing, streaming, and observability. The company also develops 1Code, an open-source desktop client for orchestrating coding agents such as Claude Code and OpenAI Codex.

Corsair is an open-source integration layer for AI agents that provides a configurable permission system for controlling agent access to external services. It stores credentials in an encrypted local database and enforces permission policies on every API call, pausing risky actions and sending review links to the user for approval. Corsair ships with pre-built integrations for services such as Slack, GitHub, Gmail, Linear, HubSpot, and Google Calendar, and supports multi-tenancy and webhook handlers.

Strix is an open-source agentic security platform that deploys autonomous AI agents to perform penetration testing on modern applications, identifying and validating exploitable vulnerabilities with proof-of-concept exploits. The platform integrates directly into development workflows, enabling continuous security testing across web applications, APIs, and codebases.

Cua provides open-source infrastructure for building and deploying computer-use agents, offering sandboxed virtual environments, an agent SDK, and benchmarking tools that enable AI agents to control full desktop operating systems including macOS, Linux, and Windows. The platform supports cloud-hosted sandboxes and a Python SDK for creating agents that can see screens, click buttons, type, and execute code autonomously.

BrowserOS is an open-source agentic browser that runs AI agents locally on your machine. It aims to automate boring tasks, facilitate expert-level research, and boost productivity by offering a privacy-first browsing experience with AI superpowers.

Memori Labs develops an open-source, SQL-native memory layer for AI agents and enterprise applications. The platform captures each conversation turn and classifies it into facts, preferences, rules, and summaries, enabling AI applications to retrieve targeted context without excessive token usage. Memori integrates with existing relational databases and offers a hosted cloud option, Memori Cloud, for teams that prefer a fully managed deployment. The system provides semantic search, relationship mapping, monitoring, and third-party integrations for bi-directional memory enrichment.

nao Labs builds an open-source analytics agent framework that enables data teams to create context-engineered agents using a file-system-based approach. The platform includes a CLI tool for building and synchronizing agent context from data warehouses, repositories, and external sources, as well as a self-hostable chat UI for deploying self-serve analytics to business users.