Autonomous & Async Agents
Agents that operate independently on tasks, often outside your local editor.
54 tools · 54 with full reviews · 25 free or open source · Updated
Jules Freemium
Google Jules is an autonomous, asynchronous AI coding agent designed to handle development tasks like bug fixing, dependency updates, and feature implementation directly within GitHub repositories. Unlike synchronous pair-programming tools, Jules operates in the background within a cloud-based virtual machine, allowing users to submit tasks and receive completed pull requests later. OpenHands Freemium
All Hands OpenHands is an open-source platform for autonomous software engineering agents that can interact with repositories, manage development workflows, and execute code. It provides a web-based control center called Agent Canvas, an SDK, and various backend options to run agents locally, in the cloud, or on private infrastructure. SWE-agent Open Source
Princeton NLP SWE-agent is a system designed to turn language models into autonomous software engineering agents capable of fixing issues in GitHub repositories. It utilizes a custom Agent-Computer Interface (ACI) to provide language models with a streamlined, interactive way to navigate, view, edit, and execute code within a environment. mini-swe-agent Open Source
SWE-agent (Princeton & Stanford) mini-swe-agent is a minimal AI software engineering agent from the team behind SWE-bench and SWE-agent. The agent class is about 100 lines of Python, it uses only bash instead of a custom tool interface, and it keeps a linear message history that is also the trajectory, so it is easy to read, modify and run with any model. Sweep Freemium
Sweep Sweep is an AI-powered coding assistant purpose-built for JetBrains IDEs that combines a low-latency custom autocomplete model with an autonomous agent for codebase-wide tasks. It integrates directly into the editor to support inline code suggestions, bug fixing, and context-aware coding assistance while maintaining user privacy. Ellipsis Freemium
Ellipsis Ellipsis is a cloud-based agent platform that allows developers to define, deploy, and manage AI coding agents using YAML configuration files. It provides isolated, secure, and governable cloud environments for agents to execute tasks, code reviews, and background workflows. Genie Freemium
Cosine Cosine is an autonomous AI software engineering agent that integrates into development workflows to plan, write, test, and submit code as pull requests. It operates through a multi-agent system designed to handle complex, long-running engineering tasks like refactors and migrations asynchronously. Open-yak Free
Open-yak OpenYak is a local-first desktop AI agent that provides a graphical interface for operating within your local file system, native desktop applications, and web browsers. It focuses on privacy by keeping conversations, files, and memory on your machine while allowing you to control and audit agent tasks in real-time. DeepAgents JS Open Source
LangChain DeepAgents JS is an opinionated agent harness designed for building autonomous agents capable of planning, file system manipulation, and subagent delegation. It is built on LangGraph, providing a production-ready runtime that supports streaming, persistence, and complex task management. Orkas Freemium
Orkas-AI Orkas is an open-source, local-first multi-agent desktop application that coordinates teams of AI specialists to execute complex projects. It uses a "Commander" to break down goals into tasks, manage agent workflows, and handle recurring work via scheduled automations. Kody Freemium
Kent C. Dodds Kody acts as an autonomous and asynchronous software factory that provides AI agents with durable memory, secure credential storage, and a cloud-based execution environment. It enables agents to turn ad hoc tasks into reusable, scheduled workflows that function independently of any specific chat interface or active session. Omnigent Open Source
Omnigent Omnigent is an open-source meta-harness that provides a unified orchestration layer for various AI agents like Claude Code, Codex, and Pi. It allows users to combine models, enforce cross-agent security policies, and collaborate in real-time via shared sessions. Manus Freemium
Monica Manus is an autonomous AI agent platform that utilizes a virtual computer to plan, research, write code, and execute complex workflows. The system features a workspace for local file interaction and supports asynchronous cloud execution, allowing tasks to continue after the user leaves the application. LoopTroop Open Source
LoopTroop AI LoopTroop is a local, open-source AI orchestrator designed for high-correctness, multi-file software engineering. It manages the full lifecycle of a coding ticket using LLM councils for planning, isolated git worktrees for execution, and human approval gates for code delivery. Muse Freemium
Meta Muse is a personal AI agent designed to perform complex, multi-step tasks and manage long-term goals by connecting with various apps and services. It functions as an autonomous assistant that can execute actions such as booking reservations, shopping, and managing small business operations, with built-in human-in-the-loop controls for sensitive steps like payments. OtoDock Freemium
OtoDock OtoDock is a self-hosted, collaborative AI agent platform that allows organizations to deploy persistent agents with individual identities, memory, and specialized tools. Built on Claude Code and Codex, it provides a secure, sandboxed dashboard for managing team workflows, automated tasks, and agent-led application development. firstmate Open Source
kunchenguid Firstmate is an agent distro that functions as a command layer to coordinate multiple autonomous coding agents across parallel tasks. It allows a single user, acting as captain, to delegate work to a crew of agents operating in isolated git worktrees and session backends, providing outcomes like completed PRs or investigation reports. OpenResearch Open Source
alphaXiv OpenResearch is a local-first research workspace that integrates with existing coding agents to turn them into research agents. It enables users to review literature, develop hypotheses, and run automated experiments while maintaining project lineage, experiment logs, and research artifacts in a local git-native structure. Coder Freemium
Coder Coder is a self-hosted platform that provides infrastructure for running AI coding agents and cloud development environments within an organization's own control plane. It enables secure, governed AI workflows by keeping the agent loop, code execution, and data within private or air-gapped infrastructure. AX Open Source
Google AX is a declarative, high-throughput orchestrator designed to run, scale, and manage autonomous agent workloads within Kubernetes clusters. It treats agents as stateful, isolated actors that can be suspended and resumed to optimize compute efficiency and resource usage. Agent Substrate Open Source
Google (donated to CNCF) Agent Substrate is a security-hardened, zero-trust runtime for running large numbers of sandboxed agents on standard Kubernetes. It maps many stateful actors onto a small pool of warm workers, snapshots idle actors out of memory and restores them on request, and keeps credentials outside the sandbox. Google donated it to the CNCF alongside gVisor in November 2026. Rivet Open Source
Rivet (YC W23) Open-source orchestrator for agentic workloads built on Rivet Actors: long-lived processes with in-memory state, a per-Actor SQLite database on S3, WebSockets and scheduling, so every agent, user, session or tenant gets its own durable process that hibernates when idle. GitHub Agentic Workflows Free
GitHub GitHub Agentic Workflows allows developers to define reasoning-based repository automations using natural language in Markdown files. The tool compiles these files into standard GitHub Actions, enabling autonomous agents to perform tasks like issue triage, code review, and documentation updates with built-in security guardrails. Strand Agents Open Source
Strand Strands Harness SDK is an open-source library for building and running AI agents in Python and TypeScript. It provides a modular, in-process framework for constructing agent loops, offering full control over tools, memory, sessions, and observability without a hosted control plane. Paperclip Freemium
Paperclip Labs Paperclip is an open-source control plane designed to manage, orchestrate, and govern fleets of autonomous AI agents as if they were human employees. It provides an organizational structure including org charts, budgets, task delegation, and approval gates to track work and costs across multiple agent runtimes. OpenSRE Open Source
Tracer OpenSRE is an open-source framework for autonomous AI SRE agents designed to investigate and resolve production incidents and CI/CD failures. It runs on your own infrastructure, allowing you to connect your observability and cloud stacks to perform evidence-backed root cause analysis and automated remediation. SuperPlane Freemium
SuperPlane SuperPlane is an open-source software factory that automates routine engineering tasks by coordinating coding agents, source control, and CI/CD tools. It evaluates backlog items for automation confidence and maintains consistent guardrails to transition tickets into verified, review-ready pull requests. Docker Agent Open Source
Docker Docker Agent is a declarative CLI tool that allows users to define, run, and share AI agents using YAML or HCL configuration files rather than writing application code. It facilitates multi-agent orchestration, tool integration, and OCI-compliant distribution for specialized AI workflows. Orbi Freemium
orbi-build (Lawrence Liu) Orbi is an autonomous AI coding agent designed to take GitHub Issues from backlog to tagged release without human supervision. It performs code implementation in isolated worktrees, followed by an independent AI-led review and test cycle before merging. Reason Freemium
Reason Machines Reason is an autonomous software engineering agent that handles coding tasks by reproducing bugs, writing fixes, running tests, and opening verified pull requests. It functions as a frontier coding harness that integrates with other agents via the Reason Network to facilitate cross-agent collaboration and task delegation. Rakazo Freemium
Elie Steinbock (elie222) Rakazo is an open-source AI agent platform that allows users to create persistent, self-hosted or cloud-based bots that can manage tasks, interact with apps, and work autonomously. Users can connect their own model subscriptions (like ChatGPT Plus or Claude Pro) or API keys, and bots can perform actions on computers, manage calendars, and collaborate with other bots. Open-CUAK Open Source
Aident AI Open-CUAK is an open-source, self-hostable kit designed for managing computer-use automation agents at scale. It allows users to run vision-based browser automation workflows locally while maintaining privacy and control. Omnara Freemium
Omnara (YC S25) Omnara is an open-source managed agent platform that provides a durable API for running, steering, and observing AI agents in production. It enables agents to operate across any model, machine, or cloud provider, while maintaining state and managing permissions through a unified control plane. Scion Open Source
Google Cloud Platform Scion is an orchestration platform designed for teams of AI agents to delegate tasks, exchange findings, and work in parallel on projects. It provides a consistent interface for executing containerized agent harnesses across environments ranging from local laptops to highly available, multi-user hosted platforms. Cindy Freemium
XD Inc. Cindy is an open-source, local-first AI agent that integrates multiple harnesses like Claude Code, Codex, and Pi into a single workflow. It allows users to execute tasks across computers and mobile devices while maintaining continuous workspace context, memory, and skills even when switching between different models or agents. Zozo HQ Freemium
Roundhouse Labs Inc. Zozo HQ is a hosted coding-agent harness that orchestrates software development by automatically dispatching GitHub or Jira issues to agent runtimes. It manages the entire lifecycle from issue intake and refinement to executing code edits in isolated environments and opening pull requests. Agensh Open Source
Microsoft Agensh is a scalable, research-oriented multi-agent harness that enables large teams of AI coding agents to collaborate without a central orchestrator. It facilitates asynchronous cooperation by using a shared Gitea workspace, a messaging interface, and a shared context mechanism. OpenBot Freemium
Nightly Labs OpenBot is a desktop and server-based platform for running teams of autonomous AI agents that work directly on your computer. It allows agents to collaborate, share files, use an embedded browser, and execute tasks using your existing AI model plans. OpenMausBot Freemium
Milind Soni OpenMausBot is an open-source, local-first alternative to Grok Bot that allows users to manage a team of AI agents directly within a messaging-style interface. It enables bots to perform tasks using the user's existing model subscriptions, offering granular control through approval cards for risky actions. AutoSwarm Open Source
HKUDS AutoSwarm is a multi-agent orchestration framework that unifies fragmented, independent AI agents into a single coordinated team. It dynamically selects agents based on their capabilities, shares context and memory across them, and manages complex task workflows with parallel execution. Lemon Code Free
Specific Labs Lemon Code is a coding agent workspace for the browser and Mac that writes code, runs it, and shows the result alongside file changes, terminal/test output, and running previews. It also connects to apps like GitHub, Slack, Linear, Notion, Gmail, and Jira to automate workflows. dsh-superagent Open Source
DeepSeek dsh-superagent is a research prototype from the DeepSeek Harness team of a proactive coding agent that serves one GitHub repository. It listens to GitHub and optionally Feishu, decides under a team-written Policy whether a request is authorized, runs each authorized request as a Ticket in its own DSH session, and replies in the original conversation, for example with a pull request. Sandcastle Open Source
AI Hero (Matt Pocock) Sandcastle is a TypeScript library and CLI from Matt Pocock's AI Hero for orchestrating AI coding agents in isolated sandboxes. A single `sandcastle.run()` call starts an agent on its own branch, sandboxes it with a chosen provider, and merges the commits back, making it easy to run parallel AFK agents and review pipelines. Agent Autoloop Open Source
Khazix Open-source local hub that turns command-line coding agents into a background loop: it takes work from webhooks, issue polling, cron rules or an HTTP API, dispatches it to Claude Code, Codex, ACP agents or any CLI, and only sends you a decision card when your call is needed. Capy Paid
Capy (YC F24, formerly Scrapybara) Capy runs a fleet of cloud coding agents in parallel, taking each task from plan to merged PR. You can start work from the Capy desktop app, the web UI, Slack, Linear, GitHub, triggers or cron. Each agent runs in an isolated cloud VM. Captains coordinate crews of agent threads, and a built-in review agent checks every PR. Boxes.dev Freemium
Boxes.dev (Nick Bushak, Drew Regitsky) Boxes.dev gives every Claude Code or Codex agent its own Linux cloud computer, a devbox cloned from a Template box of your dev environment. You start agents from Slack, Linear, a schedule or the API, and steer them from the desktop app, web, CLI or phone while your laptop is closed. Hoplite Freemium
Hoplite Hoplite runs cloud coding agents that take a task from request to reviewed code. Each thread works in an isolated sandbox with your repository, runs tests, drives a browser against the app preview and opens a GitHub pull request, then keeps iterating in the same thread as review feedback arrives. Replicas Paid
Replicas (YC P26) Replicas is a cloud coding-agent platform for engineering teams. It runs Claude Code, Codex, Cursor, OpenCode, Pi or Muse Code in isolated Linux VMs that clone your code and dependencies, started from Linear, Slack, GitHub, GitLab, the dashboard, automations or the API, and returns pull requests, replies and recordings. Tembo Paid
Tembo Tembo is an agent orchestration platform that runs Claude Code, Codex, Cursor, Amp, OpenCode, Pi and other agents in isolated cloud VMs with the full product stack. Sessions include a terminal, workspace files, a live computer and public preview links, and can be triggered from issue trackers, chat and monitoring tools. Vorflux Paid
Vorflux (YC F26) Vorflux is a cloud AI coding agent for teams that works from Slack. It plans each task, builds it on its own cloud machine with planner, builder, reviewer and QA sub-agents across several model providers, tests it and hands back a pull request with code, tests and evidence. Open Session Open Source
Tella Open Session is an open-source, self-hosted control room for coding agents built by the team at Tella. A server with a web UI and Slack, Linear, Plain and GitHub agents runs coding sessions through the Pi engine in git worktrees on your own machine or in isolated sandboxes, with multiplayer sessions and review that ends in a pull request. Nuphos Freemium
Nuphos Nuphos is an open-source shared DevOps workspace for the coding agents a team already uses. Claude Code, Codex, OpenCode, Grok Build or Antigravity run as runtimes on a managed cloud computer or your own machine, and teammates can join live sessions, share skills, tools and memory, and operate across AWS, GCP and Kubernetes with short-lived credentials and approvals. Obeya Open Source
Daniel Sadilek Obeya is a desktop workspace for directing AI coding agents on a spatial canvas. Each task is a card with its own Claude Code or Codex agent working in an isolated worktree or clone, and when the agent finishes it hands back a narrated demo that you approve or send back with feedback, by voice or text. Vespin Unknown
Stage Inc. Vespin is an inbox for all your coding agents. You describe a task, and Claude Code or Codex agents work on it in cloud sandboxes and open pull requests. Agents post a thread to your Inbox only when they need a decision, a to-do or your review, and you answer from the Mac app or the web app on your phone. FAQ
What are the best Async Agents tools?
The most complete Async Agents listings on ai.dosa.dev include Jules, OpenHands, SWE-agent. Every tool has a detail page with pricing, key features, and a verdict - see the full list of 54 tools below.
Are Async Agents tools free?
25 of the 54 Async Agents tools listed here are free or open source. Others offer freemium or paid plans - check each tool's detail page for current pricing.
How is this list curated?
ai.dosa.dev is manually curated from community submissions and editorial review. Entries are enriched with pricing, features, and reviewer notes, and the directory is updated continuously.