Agent AI apps

AI frontend agent that turns Figma, images, and prompts into production React code.

Kombai is a domain-specialized AI agent for frontend development that converts Figma designs, images, and text prompts into production-ready React and HTML. It scans your repository to match your stack and conventions, applies best practices across 400+ frontend libraries, and pairs a design canvas with code generation and in-browser refinement. On its internal 200+ task benchmark it reports a 96% compilation success rate.

Open-source integration gateway that connects AI agents to tools via MCP.

Metorial is an open-source integration and access layer for AI agents, exposing 1,200+ tools through the Model Context Protocol with consistent auth, permissions, and session observability. It can host custom MCP servers on serverless infrastructure and provides Python and TypeScript SDKs, a hosted cloud, and self-hosting. Every MCP session is recorded for review in its dashboard.

Registry and governance layer for AI agent skills.

Tessl is an agent-enablement platform that lets teams build, share, version, and govern the reusable skills their AI coding agents rely on. It pairs a searchable registry of 3,000+ public skills (plus private workspaces) with a framework to author and test skills, security scanning and policy enforcement, and usage analytics. It integrates with agents including Claude Code, Cursor, GitHub Copilot, and Gemini.

Frontend stack for building in-app AI copilots and agents.

CopilotKit is a framework for embedding AI copilots and agents directly inside web apps — chat, generative UI that renders live React components, shared application state, and frontend tool calls. It connects to agent frameworks like LangGraph, CrewAI, LlamaIndex, and Pydantic AI rather than shipping its own agent, and offers both headless APIs and prebuilt UI components. The company says its stack is used across the majority of Fortune 500 companies and powers millions of agent-user interactions per week.

Build reliable AI agents that automate UI workflows in a browser.

An Amazon model and SDK for building agents that take actions in a web browser, breaking workflows into small, testable natural-language steps for reliability at scale. Developers script and chain browser actions in Python, and run them as production UI-automation agents. Available as an AWS service after a 2025 research preview.

Enterprise web agents that run complete web workflows at scale.

Enterprise infrastructure for AI web agents that understand and act on the live web — navigating dynamic pages, filling forms, authenticating, and retrieving data even when layouts change. TinyFish runs outcome-focused workflows such as price and inventory tracking, market intelligence, and large-scale data aggregation across thousands of sites simultaneously.

Open-source infrastructure for computer-use agents across full desktops.

Cua (also styled c/ua) is infrastructure for building and scaling computer-use agents. One API boots Linux, Windows, macOS, and Android machines in lightweight, isolated virtual containers, and SDKs plus benchmarks let agents observe and control full operating systems — clicking, typing, browsing, and running real apps. On Apple Silicon it reaches near-native speed via Apple's Virtualization.framework and the Lume CLI, and it runs locally with any LLM or on Cua's managed cloud.

Local, open-source AI app builder that runs on your machine with your own API keys.

Dyad is a desktop AI app builder — like Lovable, v0, or Bolt, but running locally on your own machine. You describe an app and it scaffolds full-stack web projects, bringing your own keys across providers (OpenAI, Anthropic, Google, Ollama, LM Studio), with built-in Supabase, GitHub, and Vercel integration plus security scanning.

No-code platform to build, deploy, and manage enterprise AI agents.

StackAI is a no-code platform for designing, testing, and deploying AI agents and automated workflows that act across enterprise systems. A drag-and-drop builder wires LLMs to 100+ integrations, with human-in-the-loop oversight and model flexibility. It ships across managed cloud, VPC, and on-premise deployments.

An 'AI employee' in Slack and Teams that connects to your tools and does the work.

Viktor is an AI agent that works like a virtual coworker embedded in Slack or Microsoft Teams. It connects to 3,200+ tools, then executes tasks end-to-end — pulling data, building dashboards and reports, writing and deploying code, opening pull requests, and automating recurring workflows. It targets founders, marketing, engineering, and ops teams who want finished work delivered in chat rather than another dashboard to operate.

RL-powered agent that runs workflows across your apps and tools.

Pokee AI is an agent platform that plans, reasons, and executes multi-step tasks across your software stack from natural-language instructions. It connects via one-click OAuth to 90+ integrations (Google Workspace, Slack, Notion, Jira, GitHub, Salesforce, plus social platforms) and thousands of tools. It is built on proprietary reinforcement learning for tool selection rather than hand-built workflows, and can run hosted or fully inside your own network.

No-code 'Super Agent' that builds and runs autonomous workflow apps from plain English.

Creao AI is a no-code platform whose 'Super Agent' turns a plain-language request into a working app: it writes the logic, calls APIs, and runs the task in a sandbox. Successful workflows lock into reusable 'Agent Apps' that run on a schedule 24/7 without further AI involvement, which avoids hallucination on repeat runs. It connects to 15+ services (Gmail, Google Sheets, GitHub, Slack, Notion) and can learn new integrations from supplied API docs.

Serverless runtime and services for deploying AI agents in production.

AWS's enterprise platform for taking AI agents from prototype to production. A secure, serverless Runtime hosts agents built with any framework (Strands, LangChain, LangGraph, CrewAI) and any model, alongside modular services for Memory, Identity, Gateway (turning APIs/Lambdas into MCP tools), a Code Interpreter, Browser tool, and Observability.

Model-driven, open-source agent SDK for Python and TypeScript.

AWS's open-source SDK for building production agents with a model-driven approach: you supply a model, tools, and a prompt, and the model plans and drives the loop. Model-agnostic (Amazon Bedrock, Anthropic, OpenAI, Ollama, and more), with native MCP tool support, 20+ built-in tools, conversation/context management, hooks, guardrails, and built-in OpenTelemetry observability.

Barebones Python library for agents that think in code.

Hugging Face's minimal agent library: agents express their actions as executable Python code snippets rather than JSON tool calls, which tends to take fewer steps than tool-calling loops. Model-agnostic via LiteLLM (local Transformers, OpenAI, Anthropic, and 100+ providers), with sandboxed execution (E2B, Docker, Modal) and tools/agents shareable on the Hugging Face Hub.

Voice-native conversational AI platform for enterprise agents.

SoundHound AI builds voice-native conversational AI used to deploy autonomous agents for customer interactions across automotive, restaurants, financial services, healthcare, and smart devices. Its full-stack speech technology pairs Speech-to-Meaning understanding with the Amelia enterprise agent platform and newer OASYS stack, automating phone answering, drive-thru ordering, and IT service management. It powers billions of conversations a year and is offered to developers and enterprises via APIs and embedded SDKs.

An autonomous AI SRE that triages alerts and finds root causes in minutes.

Traversal is an AI site reliability engineer (SRE) that autonomously triages noisy alerts, traces incidents across services and dependencies to isolate root causes, and feeds production context back to prevent recurrence. Its causal search engine and 'production world model' narrow failures down at petabyte scale. At American Express it reached 82% root-cause accuracy and cut potential MTTR by ~32%, with PepsiCo also among its enterprise users.

Self-learning AI agents for enterprise operations.

Beam AI is an agentic process automation platform that turns written standard operating procedures into self-learning AI agents for back-office work. Agents handle data entry and extraction, document processing, and communication tasks across functions like finance, HR, and customer operations, integrating with existing enterprise systems. Beam targets mission-critical workflows where data privacy and deep integration matter, and reports production use at large enterprises.

Enterprise RAG platform for building grounded, specialized AI agents over technical knowledge.

Contextual AI is an enterprise platform for building retrieval-augmented AI agents that answer over a company's technical documentation, specifications, and institutional knowledge. It exposes individually tunable RAG components — document parsing, retrieval, reranking, and grounded generation — plus an agentic search and research layer, so teams in regulated, high-accuracy domains (finance, engineering, legal) can automate complex knowledge work while keeping answers grounded in source material. Available as multi-tenant SaaS, single-tenant SaaS, or private VPC, with developer APIs.

Open-source toolkit to connect any LLM to any MCP server.

mcp-use is an open-source framework for building MCP clients, servers, and agents in TypeScript and Python. It lets you connect any LLM to any Model Context Protocol server in a few lines of code — configuring servers, choosing a model, and spinning up a tool-using agent — without depending on a closed application client. On top of the official SDK it adds opinionated developer ergonomics: hot reload, dev tools, auth providers, and UI components.

Open-source AI coding agent that lives in your terminal.

Qwen Code is a command-line coding agent from Alibaba's Qwen team, adapted from Google's Gemini CLI and tuned for the open-weight Qwen3-Coder models. It runs an interactive or headless terminal UI with agentic features — auto-memory, subagents, agent teams, and MCP support — for codebase comprehension, editing, and multi-step automation. It supports OpenAI-, Anthropic-, and Gemini-compatible providers plus local models, and offers a free Qwen OAuth tier alongside bring-your-own-key usage.

Open-source, code-first framework for building and deploying multi-agent systems.

Agent Development Kit (ADK) is Google's open-source framework for building, evaluating, and deploying AI agents and multi-agent systems with a code-first, software-engineering workflow. Agents are defined in code and composed into hierarchies, equipped with prebuilt tools, MCP tools, third-party libraries, or other agents. ADK supports bidirectional audio and video streaming, ships built-in evaluation, and deploys anywhere — your own infrastructure, Cloud Run, GKE, or Vertex AI Agent Engine.

Source-available platform for building AI agents for support and docs.

Inkeep is an AI agent platform for customer-experience and operations teams, pairing a no-code visual builder with a TypeScript framework that stay in two-way sync. Agents draw on unified search and RAG over your docs and company data, connect to apps via MCPs and integrations, and report monitoring insights. The Inkeep Agents framework is source-available on GitHub and self-hostable, alongside a managed cloud.

Framework and browser platform for building reliable AI web agents.

Notte is a full-stack platform for building and running AI web agents: it turns any website into a structured action API so agents can click, fill forms, and navigate flows instead of scraping raw text. It offers managed cloud browser sessions, serverless browser functions, credential vaults, proxies, CAPTCHA handling, session replays, and observability, with compatibility for Playwright, Puppeteer, Selenium, and CDP tooling. A source-available framework on GitHub underpins the hosted platform.

Cloud headless-browser infrastructure for AI agents.

Managed cloud platform that gives AI agents scalable headless browser sessions with built-in stealth mode, proxy rotation, and CAPTCHA solving. Sessions launch in under a second and are driven via API with Playwright, Puppeteer, or Selenium. It also offers HyperAgent, an open-source framework that extends Playwright with natural-language commands, and an official MCP server so LLMs like Claude can browse the live web.

Agentic AI workspace on an infinite canvas with autonomous multi-task agents.

Flowith is an AI workspace built around an infinite two-dimensional canvas, where you branch threads and organize work visually instead of in a linear chat. Its autonomous agent, NEO, plans and executes multi-step tasks with tool calling and persistent context, while the Oracle scheduler breaks complex jobs into subtasks. It supports leading models including GPT and Claude. A free Starter tier offers limited credits, with paid Professional, Ultimate, and Infinite plans.

Agentic AI platform for IT, HR, and customer service desks.

Aisera is an enterprise agentic AI platform that automates service interactions across IT, HR, finance, and customer support. It resolves and deflects tickets autonomously using domain-specific 'Hyperflows' and prebuilt templates, with tools like Agent Composer and LLM Studio to configure conversational and back-office automation in one loop.

Salesforce's platform for building and deploying autonomous AI agents.

Agentforce is Salesforce's platform for building, customizing, and deploying autonomous AI agents that act on enterprise data. Prebuilt agents handle service, sales, and other workflows, grounded in Salesforce CRM data and Data Cloud and reasoned over by the Atlas Reasoning Engine with enterprise guardrails. The 2025 Agentforce 360 release expanded it into a platform for orchestrating humans and agents together, with tools like Agent Script for custom agent logic.

Chat with AI agents that design, code, test, and deploy your app.

Emergent is a vibe-coding platform that turns a conversation into a production-ready app — a coordinated team of AI agents designs, builds, tests, and deploys full-stack web and mobile applications from natural-language prompts. It targets small businesses and non-developers shipping software without a codebase.

Build, deploy, and govern company AI agents as code.

Kortix is an open-source platform for building internal AI agents — defining their skills, tools, personas, and guardrails as code, running them in isolated sandboxes that propose work via git branches and pull requests. It ships as the open-source Suna agent plus a managed cloud, with self-host, on-prem, and air-gapped deployment options.

An AI 'super agent' office suite that drafts docs, slides, sheets and more.

Skywork is an agentic AI office suite that generates editable documents, slideshows, spreadsheets, webpages and podcasts from a prompt, backed by a deep-research pipeline. It pairs five specialized 'expert' agents — for docs, slides, sheets, podcasts and webpages — with a general agent for multimodal content. It launched globally in May 2025.

AI code review and stacked pull requests for fast-moving engineering teams.

Graphite is a code-review platform built around stacked pull requests and Diamond, an AI reviewer that comments on PRs with senior-engineer rigor — catching bugs, style issues, and architectural concerns. It layers a merge queue, a CLI, a VS Code extension, and a collaborative review interface on top of GitHub-based workflows, and now runs agents directly inside the PR.

Open SDK for production AI agents in .NET and Python.

Microsoft's multi-language framework for building, orchestrating, and operating AI agents and multi-agent workflows. It offers a consistent API across C#/.NET and Python with first-class support for both, plus built-in durability, restartability, observability, and human-in-the-loop control. Provider-flexible, so the underlying models can change without rewrites.

Visual low-code builder for AI agents and RAG apps.

Open-source tool for prototyping and deploying agentic and RAG applications through a drag-and-drop canvas, with Python customization underneath each node. It connects major LLMs, vector databases, and tools, and can expose flows as APIs or MCP servers. Run it locally, as a desktop app, or on managed cloud.

Open-source Python framework for production RAG and agents.

Orchestration framework for building LLM applications as modular pipelines — retrieval, routing, memory, and generation wired together with explicit, traceable data flow. It is model- and store-agnostic, integrating major providers and vector databases behind a stable component API. Aimed at production: serialization, logging, and deployment across cloud or on-prem.

Open-source autonomous coding agent inside VS Code.

VS Code extension that turns the editor into an agentic coder with real file-system access, terminal control, and multi-step task execution. It ships role-specific modes — Architect, Code, Ask, Debug, Test — each with its own prompt and tool permissions, and can hand off between them mid-task. Model-agnostic: connect any provider key, from frontier APIs to a local model, and pay only for your own token usage.

No-code platform to build, launch, and scale AI agents.

Voiceflow is a no-code platform for designing, testing, and shipping chat and voice AI agents for customer support, lead generation, and beyond. A visual canvas lets designers, PMs, and developers collaborate on conversation logic, connect a knowledge base for RAG, and deploy to web widgets, phone, or any channel via API. It supports multiple model providers and scales to enterprise call and message volumes.

Build native mobile apps by chatting with AI.

Rork turns plain-language prompts into working iOS and Android apps. Its Pro tier generates React Native + Expo projects you can preview live and ship to the App Store and Google Play, while Rork Max builds native SwiftUI apps compiled with Xcode. It focuses on mobile as the target surface rather than web pages, and produces real native apps instead of web wrappers.

Agentic legal AI that drafts, reviews, and negotiates contracts from your playbooks.

Genie AI is a legal AI platform for drafting, reviewing, editing, and negotiating business contracts. It can generate personalized, jurisdiction-aware agreements — up to 50+ pages — in minutes, flag risks in existing documents, and learn an organization's standards and playbooks to keep edits on-policy. London-based, it offers a free tier plus paid Pro, Business, and Enterprise plans for in-house legal teams and firms.

No-code platform to build and deploy AI agents anywhere.

MindStudio is a no-code platform for designing, building, and deploying AI agents. You assemble agents visually and ship them to web apps, browser extensions, email triggers, scheduled automations, APIs, and MCP servers, with optional custom JavaScript or Python for advanced steps. It connects 1,000+ apps and gives access to 200+ AI models through one interface, charging model usage at provider cost with no markup.

Build and share AI mini-apps from a plain-language description.

Opal is an experimental Google Labs tool for building AI mini-apps without code. You describe a task in natural language and Opal generates a workflow that chains prompts, models, and tools, which you can refine in a visual node editor and share with others via a link. It targets marketers, analysts, and other non-engineers who want to prototype and ship small AI-powered tools quickly.

High-performance Python framework for multi-agent systems.

Agno is an open-source Python framework for building multi-agent systems with memory, knowledge, tools, and reasoning. Agents are fast to instantiate and can be composed into teams, and the companion AgentOS runtime serves them in production as a FastAPI app with a control plane.

Build and deploy AI agents as APIs, with MCP tools built in.

Toolhouse is a backend-as-a-service for AI agents: define an agent in plain language or code and deploy it as an API in one command. Agents connect automatically to the Toolhouse MCP server, which provides RAG, memory, code execution, browser use, and 1,000+ tool integrations over the Model Context Protocol.

Open-source framework and cloud for realtime voice, video, and physical AI agents.

LiveKit is the realtime infrastructure layer most voice-AI stacks are built on. Its open-source Agents framework wires together streaming speech-to-text, an LLM, and text-to-speech with reliable turn detection, interruption handling, and telephony, so agents can join a session and converse in near real time. You bring your own STT, LLM, and TTS providers; LiveKit handles the low-latency WebRTC transport. It runs self-hosted under Apache 2.0 or as a managed cloud platform.

Open-source AI coding agent for VS Code, JetBrains, and the CLI.

Kilo Code is an open-source AI coding agent that plans, writes, edits, and runs code inside your editor. It offers specialized modes — Code, Architect, Ask, Debug, and Orchestrator — works across VS Code, JetBrains IDEs, and the CLI, and supports 500+ models with bring-your-own-key and zero markup. A free tier plus paid cloud and team options round it out.

Run a team of parallel AI coding agents on your Mac.

Conductor is a macOS app that runs multiple AI coding agents — Claude Code, Codex, and Cursor — in parallel, each in its own isolated git worktree. You dispatch tasks, watch each agent's progress in real time, then review and merge their changes without them stepping on each other. The app itself is free and runs on your own agent subscriptions; Conductor Cloud extends runs beyond your laptop.

Charm's glamorous agentic AI coding agent for your terminal.

A terminal-based AI coding agent from Charm, the team behind Bubble Tea and other CLI tools. It plans and edits files, runs commands, and pulls code context via LSP, all inside a polished TUI. It is model-agnostic: bring your own keys for Anthropic, OpenAI, Google, Groq, OpenRouter, and other providers, and switch models mid-session.

Multi-model AI super-assistant with an agent layer for tasks and apps.

An AI platform whose consumer product, ChatLLM, puts many frontier models — GPT, Claude, Gemini, and others — behind one subscription, alongside image generation and a code/app builder. Its DeepAgent layer runs multi-step agentic tasks like dashboards, automations, and app builds. The company also sells an end-to-end enterprise generative-AI and ML platform.

Describe an app in plain English and get a working full-stack product.

An AI no-code app builder that turns a conversational prompt into a fully functional web app — database, user auth, permissions, and hosting handled for you. It targets non-developers and founders who want to ship custom internal tools and products without writing code. Now operated by Wix following its 2025 acquisition.

Enterprise platform for building and sharing custom AI agents on company knowledge.

An enterprise AI platform where teams build custom agents over shared company knowledge and reuse them org-wide. Connectors pull in Slack, Notion, Google Drive, and other workspace tools; each agent can run on a choice of frontier models. Used by 3,000+ organizations with over 300,000 agents deployed.

Claude Code's agentic harness, repointed at non-coding knowledge work.

Anthropic's agentic desktop mode that completes multi-step knowledge work end-to-end. Point it at local folders and connected apps and it organizes files, builds spreadsheets, and prepares reports — showing its plan for approval before significant actions. Launched as a research preview in January 2026, it is now included in every paid Claude plan on macOS and Windows.

AI code review with full-codebase context on every pull request.

Greptile builds a graph index of your repository and runs a swarm of agents over each pull request, catching cross-file logic bugs, security risks, and style violations that diff-only reviewers miss. It learns team standards from past review comments, enforces custom rules written in plain English, and reviews PRs on GitHub and GitLab, with TREX generating and running tests per PR.

OpenAI's coding agent across terminal, IDE, web, and ChatGPT.

Codex is OpenAI's agentic coding tool that reads, edits, and runs code on your behalf. It spans a local terminal client (the open-source Codex CLI), a VS Code extension, a cloud agent that executes tasks in isolated containers with full repo access, and a surface inside ChatGPT. It runs on OpenAI's Codex-tuned models and can parallelize multiple agent tasks at once.

The open-source, provider-agnostic AI coding agent for your terminal, IDE, or desktop.

OpenCode is an open-source AI coding agent from Anomaly, the team behind SST. It runs in the terminal, a desktop app, or an IDE extension, connects to 75+ model providers through Models.dev — frontier APIs and local models alike — and stores none of your code on external servers. The optional paid Zen gateway adds a curated, benchmarked model set under one bill.

AI platform for plaintiff law firms — works whole cases from intake to resolution.

Legal AI built specifically for plaintiff-side law firms in areas like personal injury and employment law. Eve evaluates incoming cases, builds medical chronologies, drafts complaints, demand letters, and discovery, and runs an AI Auditor that scans the firm's entire active caseload nightly to surface missed opportunities. The company says it is used by over 1,000 plaintiff firms and processes more than 200,000 cases a year.

Autonomous AI agents built for accounting firms.

Basis builds AI agents purpose-built for accountants that execute complex accounting workflows end-to-end and return finished output ready for review. It integrates with existing ledgers and systems to handle tax, audit, advisory, and client-accounting tasks, adapting to each firm's processes rather than relying on rigid rules.

Open-source MCP server giving AI clients terminal and file control.

Desktop Commander is an open-source Model Context Protocol server that lets MCP clients like Claude Desktop, Cursor, and VS Code run terminal commands, search the file system, and make surgical diff-based edits. It manages long-running processes, reads and writes files (including Excel and PDF), and turns an AI assistant into a hands-on coding and automation agent on your own machine. Free and MIT-licensed, with one of the larger community followings among MCP servers.

AI workspace for tasks, projects, and multi-agent automated workflows.

Taskade turns prompts into apps, AI agents, and automated workflows in one workspace, blending task lists, projects, notes, and mind maps with real-time collaboration. Its AI agents handle research, writing, and reviewing with multi-agent handoffs, plus durable automations and 100+ integrations.

Domain-specific AI for legal and professional services.

Harvey is an AI platform built for law firms and corporate legal teams to research, analyze documents, and draft. It pairs an assistant and a secure document vault with purpose-built agents that execute complex legal work end to end. Sold through enterprise contracts rather than self-serve.

Open-source SDK for building reliable AI browser agents.

Stagehand is an SDK from Browserbase for building browser agents in TypeScript or Python. It exposes act(), extract(), and observe() primitives so you can drive a page with natural language while keeping deterministic code wherever you need it. v3 dropped its hard Playwright dependency for a modular Chrome DevTools Protocol driver.

Open-source visual builder to create, deploy, and orchestrate AI agents.

An open-source workspace for building AI agents on a Figma-like drag-and-drop canvas, conversationally, or in code. It connects LLMs to 1,000+ integrations plus knowledge bases and structured tables, then deploys a workflow as an API, scheduled job, webhook handler, or standalone chat app. Apache-2.0, YC-backed, and runnable in Sim's cloud or self-hosted via npm or Docker.

Agentic AI that automates document-heavy knowledge work and data extraction.

An operational AI platform from V7 Labs that builds and runs agents over complex documents — extracting financial, legal, and commercial terms, completing DDQs, and generating memos with source traceability. It chains foundation models from OpenAI, Anthropic, and Google into multi-step, auditable workflows aimed at finance, insurance, legal, and real-estate teams.

Open-source crawler that turns the web into clean, LLM-ready Markdown.

Crawl4AI is an open-source (Apache 2.0) web crawler and scraper built for AI pipelines, converting pages into clean Markdown or structured data for RAG, agents, and data pipelines. The core runs locally with no API key, handles JS rendering, and supports optional LLM-based extraction with any provider. It installs as a Python library/CLI or deploys as a Dockerized FastAPI server; a hosted Cloud API is in closed beta.

Automate browser-based workflows on any website with AI.

An AI agent that completes browser workflows — form fills, logins, data extraction, multi-step flows — by combining computer vision with LLMs rather than hand-written selectors, so a single agent generalizes across sites it has never seen. Run it via the cloud app and API or self-host the open-source engine; bring your own model (OpenAI, Anthropic, Gemini, or local Ollama).

Continual learning memory for stateful agents. Better context, fewer tokens.

A memory and user-personalization layer for AI agents that keeps reasoning about each user across sessions, so apps get richer context without stuffing whole histories into the prompt. It models peers and sessions, runs background inference to derive durable facts, and answers natural-language questions about a user at query time. Available as a managed API or self-hosted FastAPI server, with Python and TypeScript SDKs.

The MCP runtime for secure, authenticated agent tool-calling.

A platform that lets AI agents take real, authenticated actions across business systems like Gmail, GitHub, Slack, and Notion. Arcade handles the hard part of agent tool-use — secure, per-user authorization — so a tool call runs with the actual end user's permissions rather than a shared API key. It ships agent-optimized prebuilt tools plus an open framework for building custom ones, exposed over MCP. A free tier is available for getting started.

Spec-driven agentic IDE — turn prompts into specs, then code.

Kiro is AWS's agentic IDE that brings engineering rigor to AI coding. Instead of vibe-coding straight to a diff, it first generates durable spec artifacts — requirements, a technical design, and a task list — then implements them. Agent hooks automate routine actions on file events, and it runs on Claude models.

Privacy-first AI coding assistant you control — completions, chat, and agents.

An AI coding assistant built for enterprise control, offering completions, chat, and agentic workflows across all major IDEs. It never trains on or retains your code, and can be deployed as SaaS, in a VPC, on-premises, or fully air-gapped. Supports access to major LLMs plus bring-your-own and private models.

All-in-one AI workspace with an autonomous Super Agent.

An AI workspace built around a Super Agent that breaks goals into sub-tasks, picks the right tools, and finishes work autonomously — building slides, sheets, docs, websites, and videos, and even placing phone calls. It orchestrates multiple large language models and dozens of integrated tools, synthesizing results into shareable Sparkpages. Built by a team from Google and Microsoft and available on web and mobile.

Authenticated toolkits that give AI agents access to SaaS apps, via MCP or API.

Tool layer that gives AI agents secure, managed access to 1000+ toolkits and 20,000+ tools across SaaS apps, exposed via MCP or direct APIs. It handles authentication, tool search, and context management, and its MCP Gateway hands each team a managed endpoint to paste into Claude, Cursor, or ChatGPT. The core is MIT-licensed; a free tier covers 20K tool calls per month.

AI code review and test generation across IDE, PRs, and CLI.

Qodo (formerly CodiumAI) is a code-quality platform that runs AI review on pull requests, generates tests, and assists inside the editor. Its multi-agent reviewer pulls context from the codebase and PR history across GitHub, GitLab, Bitbucket, and Azure DevOps, with VS Code and JetBrains plugins plus a CLI so issues surface before merge.

AI code review on every pull request, IDE, and CLI.

CodeRabbit posts contextual, line-by-line AI reviews on pull requests across GitHub and GitLab, with an agentic chat for follow-ups plus integrated linters and SAST tools. The same reviewer runs in the IDE (VS Code, Cursor, Windsurf) and from a CLI, so issues surface before the PR.

Open-source browser automation for AI agents.

MIT-licensed Python framework that connects AI agents to a real browser to navigate, fill forms, extract data, and run multi-step web workflows. It parses the live DOM into a structured view so any model — Claude, GPT, Gemini, Qwen, DeepSeek — can act on it. A hosted Browser Use Cloud adds managed sessions and a fully-hosted agent.

Google's open-source Gemini AI agent, in your terminal.

An open-source command-line agent that brings Gemini into the terminal for coding, debugging, content generation, and research. Ships built-in Google Search grounding, Model Context Protocol support, and project context via a GEMINI.md file, plus a non-interactive mode for scripting and CI. Free to use by signing in with a personal Google account, with paid options through AI Studio, Vertex AI, and Gemini Code Assist.

No-code AI workflow automation on a visual node canvas.

A drag-and-drop platform for building AI-powered automations as node graphs — web scraping, document parsing, image analysis, and multi-model chaining without code. Connects apps, data, and LLMs into triggered or scheduled flows. Backed by Y Combinator and a Benchmark-led Series B.

Full-stack enterprise generative AI platform for building agents and apps.

An enterprise platform that pairs an application layer for building AI agents and workflows with its own foundation models, the Palmyra LLM family. It targets regulated industries with SOC 2 Type II, PCI, and HIPAA compliance and a no-training-on-customer-data policy. Specialized models like Palmyra Med and Palmyra Fin tune the platform for healthcare and finance use cases.

OpenAI's open-source coding agent for the terminal.

A lightweight, open-source coding agent that runs locally in your terminal, reading your repository, editing files, and running commands in a conversational loop you review in real time. Built in Rust for speed, it supports Model Context Protocol tools, subagents for parallel tasks, and switching between GPT models. Sign in with a ChatGPT plan (Plus, Pro, Business, Edu, or Enterprise) or use an OpenAI API key.

Async coding agent that clones your repo and ships PRs.

Jules is Google's asynchronous coding agent. It clones your GitHub repository into a secure Google Cloud VM, understands the full project context, then writes tests, fixes bugs, and makes multi-file changes in the background before opening a pull request. A Jules Tools CLI and a Jules API extend it into your own workflows.

AI pair programmer in your editor — completions, chat, and agents.

The original in-editor AI coding assistant, offering multi-line completions, chat with codebase context, agent mode, and a cloud coding agent that opens pull requests. Runs across VS Code, JetBrains, Visual Studio, the CLI, and GitHub.com. A free tier covers light use; paid tiers add premium models and more allowance.

Free AI-native IDE with an autonomous build mode.

VS Code-based AI IDE from ByteDance whose Builder/SOLO agent plans tasks, edits across files, runs commands, and previews results. Bundles access to frontier models like Claude, GPT, Gemini, and DeepSeek without requiring your own API key, on free and paid usage tiers.

Agent-native software development with autonomous 'Droids'.

An agent-native platform for software development built around 'Droids' — autonomous coding agents that plan, edit, and run tasks across a codebase. Works from a CLI/SDK and runs agents both in the cloud and locally in the background. Model-agnostic across frontier and open-weight models.

Agentic coding tool from Sourcegraph for terminal and editor.

Amp is an agentic coding tool from Sourcegraph that works in the terminal and across editors, autonomously editing code and running tasks. It routes between frontier models across deep, smart, and fast modes, and centers on shareable 'threads' so teammates can view and search each other's agent sessions.

Headless browser infrastructure for AI agents.

Managed cloud fleet of headless browsers that let AI agents browse, authenticate, and act on the web at scale. Sessions ship with stealth proxies, automated CAPTCHA solving, and observability, driven via API or the open-source Stagehand framework. Usage-based billing on top of a monthly base plan.

Type-safe Python agent framework, the Pydantic way.

An open-source Python framework for building production-grade agents with validated, structured outputs instead of raw-string parsing. Model-agnostic across OpenAI, Anthropic, Gemini and many more, with composable tools, durable execution, MCP support, and built-in observability via Pydantic Logfire. MIT-licensed from the team behind Pydantic.

Natural-language interface that lets LLMs run code locally in your terminal.

Gives LLMs a ChatGPT-like terminal interface to write and execute Python, JavaScript, Shell, and more on your own machine, asking approval before each run. It works with hosted models or fully local models via Ollama, LM Studio, or Jan using a --local flag. Free and open source under AGPL-3.0; you supply your own model (API key or local).

Moonshot's assistant — long-context chat, deep research, and agents.

Moonshot AI's assistant — long-context chat, Deep Research, and an agent mode (Kimi Code, parallel subagents). Built on the open-weight Kimi K2 model family.

Visually build AI agents and LLM workflows — drag-and-drop, self-hosted.

A low-code visual builder for AI agents, chatflows, and multi-agent systems on a node canvas. It self-hosts locally through npm or Docker, or deploys to major clouds, and is provider-agnostic across many LLMs via its node ecosystem. The core is Apache-2.0; a managed Flowise Cloud tier exists, making it freemium.

Cloud IDE + Agent that builds, runs, and deploys from a prompt.

A browser IDE with Replit Agent — describe an app and it scaffolds, codes, runs, and deploys it, with a hosted database and one-click publish. Zero local setup, ships from anywhere.

Self-improving personal AI agent that learns skills and keeps persistent memory.

A self-hostable AI agent from Nous Research that builds skills from experience, maintains long-term memory, and is reachable through a terminal TUI, a web dashboard, or messaging gateways (Telegram, Discord, Slack). It runs anywhere from a small VPS to local Docker, and works with any model via Nous Portal, OpenRouter, OpenAI, or self-hosted endpoints. Free and open source under MIT; you bring your own model access.

On-machine AI agent for coding, workflows, and automation.

An extensible AI agent from Block that runs natively on your machine as a desktop app or CLI, executing code and automating engineering tasks via MCP extensions. It supports 15+ LLM providers (Anthropic, OpenAI, Google, Ollama, and more) using your own keys. Free and open source under Apache 2.0.

Open-source AI coding assistant for VS Code, JetBrains, and the CLI.

An open-source AI coding assistant delivered as a VS Code extension, a JetBrains plugin, and a CLI, with chat, autocomplete, and agent workflows. It runs locally in your editor and works with any model provider, including local models via Ollama, using your own keys. The core is Apache-2.0 and free; the company also offers a hosted Continue Hub for sharing and managing assistants.

The data framework for LLM apps — RAG, agents, and document workflows.

An open-source framework (Python + TypeScript) for connecting LLMs to your data — ingestion, indexing, retrieval, and agentic document workflows. Pairs with the managed LlamaCloud (LlamaParse) for production parsing and extraction. The most-used RAG framework after LangChain.

Agentic development environment born out of the terminal.

A modern terminal and agent platform — multi-agent orchestration, codebase indexing, a built-in editor (Warp Code), and granular permission controls for running coding agents. Adds the Oz cloud layer for remote agent execution and team knowledge via Warp Drive. Free tier with monthly AI credits; paid Build/Max/Business plans add more credits and BYOK.

Prompt a full-stack app into existence — UI, backend, and deploy.

An AI app builder that turns prompts into full-stack web apps — React UI, Supabase backend, GitHub sync, and one-click publish. Popular for going from idea to shipped product fast.

Visual platform for agentic workflows, RAG pipelines, and LLM apps.

An LLMOps platform that bundles a drag-and-drop workflow builder, RAG pipelines, agent tooling, model management, and observability into one surface — prototype to production without much glue code. Connects hundreds of proprietary and open models across providers. Self-host the source-available edition for free, or use Dify Cloud's paid tiers.

All-in-one private AI app for chatting with your documents, with agents.

An all-in-one application for private, ChatGPT-style chat over your own documents, with built-in RAG, AI agents, and multi-user workspaces. Runs as a local desktop app (Mac/Windows/Linux) or self-hosted via Docker, and supports 40+ LLM providers plus local models with your own keys. Open source under MIT; Mintplex Labs also offers a paid hosted instance, making it freemium.

Autonomous AI agent that runs deep multi-source web research and writes cited reports.

An open-source autonomous research agent that plans a task, runs parallel multi-source web searches, validates sources, and synthesizes a cited report. It runs locally as a Python package, FastAPI server, or MCP server, and works with any LLM provider plus a search/retriever backend. Free and Apache-2.0 licensed — you only pay your own LLM and search-API costs.

Agentic automation. Long-running agents instead of one-shot triggers.

Builds AI agents that wait for emails, calendar invites, or webhooks — then take multi-step actions on your behalf. The 'agentic Zapier' pitch; strong for ops workflows that need judgement, not just routing.

Multi-agent framework with explicit roles and tasks.

Python framework for orchestrating crews of specialised agents — researcher, writer, reviewer — coordinated through shared context. Opinionated about roles, sequencing, and delegation; good fit for content-and-research pipelines.

Autonomous software engineer agent. Plans, codes, ships from a prompt.

Cognition's hosted SWE agent. Operates inside its own sandbox with a browser, terminal, and editor — plans the work, writes the code, and submits a PR. Aimed at long-horizon tickets rather than tab-completions.

Prompt-to-UI generator from Vercel. Outputs React + Tailwind + shadcn.

Vercel's generative UI surface. Type a description, get a working React component using Tailwind and shadcn primitives — copy the code or fork into a v0 project for iteration.

Graph-based agent orchestration. Stateful loops with checkpoints.

LangChain's agent layer. Model agents as nodes in a state graph with persistent checkpoints, human-in-the-loop steps, and durable execution. Strong when you need a long-running, debuggable agent rather than a one-shot chain.

Build vision detectors and agents from a few labeled examples.

Build vision applications with a labelling-light workflow — point at examples, get a deployable detector. Recently extended into vision agents that reason over images and PDFs without bespoke training.

Anthropic's official CLI / IDE / web agent.

Anthropic's agentic coding tool. It reads the spec, drafts the diff, runs the tests, and opens the PR — across the CLI, IDE extensions, the desktop app, and the web.

Open-source autonomous SWE agent. Successor to OpenDevin.

Self-hostable agent platform with a browser, terminal, and editor in a sandbox — comparable to Devin in shape, open in licence. Active research community pushing the autonomous-SWE bar.

Prompt-to-app generator running in the browser. WebContainers under the hood.

StackBlitz's in-browser full-stack app builder. Generates a working project — frontend, backend, deps — and lets you iterate in a sandboxed Node runtime without leaving the tab.

Open-source autonomous coding agent in VS Code. BYO key.

Community-maintained VS Code extension that turns the editor into an agent surface — reads the repo, runs commands, edits files, and verifies its work. Pairs naturally with Claude. Formerly known as Claude Dev.

The default open-source framework for composing LLM apps.

Python + TypeScript framework for chaining prompts, tools, retrievers, and memory into LLM applications. Ubiquitous in the ecosystem; pairs with LangGraph for agent orchestration and LangSmith for tracing.

General-purpose autonomous agent for research, ops, and content.

Cloud agent that browses, drafts, schedules, and follows multi-step instructions over hours. Closer to a virtual operator than a coding tool — strong for non-code knowledge work. Acquired by Meta; product continues to ship.

Terminal-native pair programmer. BYO key, BYO model.

Open-source CLI for AI pair-programming directly against your git repo. Brings any model (Claude, Gemini, GPT, local) into a tight commit-per-task loop.