Coding Agents

Reason is the autonomous AI software engineer and cloud coding agent. Connect GitHub, assign issues,

Reason is an autonomous AI software engineer and cloud coding agent. It works on software tasks such as investigating bugs, editing code, running tests, and merging pull requests. The page shows Reason coordinating work through subagents, benchmarking and testing code changes, building dashboards, and responding to requests in Slack. It also describes starting tasks from a phone and checking on them anywhere.

Run coding agents in parallel. Step in only when one needs you.

Natamux is a free, MIT-licensed desktop app for running Claude Code, Codex, Gemini CLI, GitHub Copilot, Cursor and other CLI agents in parallel. Each task gets its own Git worktree and terminal, with stages from specification through pull request managed on one dashboard. Developers can approve plans and test specifications, review agent work, track tasks in a table or board, and manage GitHub changes without leaving a task. The app runs on macOS, Windows and Linux and requires no account.

Take apart any app and ask your agent what it does. Agent-first reverse engineering for Mac, Windows

Eleven is a desktop app for examining software binaries and asking an AI agent about them. It supports Mac, iOS, Linux and Windows programs, Android apps, GPU shaders and ML models, with views including pseudocode, assembly, control-flow graphs and hex. It is for security researchers, developers checking dependencies, and people curious about how apps work. Agents can inspect functions and suggest changes; edits require approval when review is enabled, and byte patches are written to a copy rather than the original. Eleven runs locally, while anything the agent reads goes to that agent’s provider.

Das KI-Interface für Ihren Online-Shop

Lotse ist ein KI-Interface für Online-Shops, das die Pflege von Produkten, Preisen, SEO und Landingpages über einen Chat ermöglicht. Änderungen werden stets im Vorher-Nachher-Vergleich angezeigt, und das Schreiben erfolgt erst nach Ihrer Freigabe. Es unterstützt Plattformen wie Shopware, Shopify, Magento 2 und OpenMage und ist quelloffen.

Run agentic teams from your own computer.

Bees is an open-source AI platform that allows you to run autonomous agents on your own computer. It keeps your files, sessions, passwords, and data local, ensuring privacy and control. Users can utilize local models or integrate existing subscriptions with no additional fees per token.

A single Rust binary that drives the WebView your OS already ships.

Navette is designed to provide a lightweight browser engine for AI agents, utilizing the native web rendering capabilities of your operating system without including bulky dependencies like headless Chromium. It operates through a simple binary, which can run across various platforms including macOS, Windows, and Linux. Its small size, typically around 1 MB, allows quick installation and setup, making it ideal for developers looking to automate web interactions efficiently.

Two-way code review between you and Claude Code, inside VS Code.

Redline facilitates a collaborative code review process between you and Claude Code directly within Visual Studio Code. By utilizing comment threads on diff lines, Claude enables seamless dialog as you review code changes together. This integration is designed for developers who seek an efficient and interactive review experience, ensuring every line is accounted for and feedback is easily provided.

Share Claude Code, Codex, OpenCode, Gemini CLI, Copilot CLI, Cursor and ZCode sessions with your tea

Session Relay helps teams share coding-agent sessions and hand work off between teammates, including across different coding agents. A share includes a handoff with the goal, decisions, completed work, next steps and open questions, plus recent messages and files touched. Teammates can find shared sessions in the web app, CLI or an agent and continue with `/pickup`. It supports team features such as roles, repo-specific handoff templates, retention settings, audit logs, comments and handoff chains. Session Relay is open source and can be self-hosted with Docker or on Cloudflare Workers; its hosted cloud offering is listed as coming soon.

Sakura Lang is an agent-first programming language. Every signature declares its effects, errors and

Sakura Lang is a programming language and compiler for code written by AI agents. Function signatures declare effects, errors and contracts, which the compiler checks; agents can also work with structured diagnostics, typed holes and fixes that state their impact. The compiler includes checking, verification, testing, building, formatting, an LSP server and MCP tools. The alpha.1 preview supports JavaScript and WebAssembly targets and is available through npm, Homebrew and install.sh.

Secure, scalable software — engineered across product, security and AI.

Indigloo Softwares builds enterprise software, provides cybersecurity services, and develops AI solutions. Its work spans web and mobile applications, APIs, vulnerability assessment and penetration testing, AI agents, and workflow automation. The company works with organizations from discovery and strategy through engineering, testing, deployment, and ongoing support. Its cybersecurity services cover web, API, and infrastructure layers, with remediation guidance for client teams.

Bots de IA com navegador, terminal e agenda, trabalhando por você com a sua chave.

Touri is a platform for running AI bots with their own browser, terminal, repositories, and task schedule. Users can chat with bots, watch their work live, and manage recurring tasks, GitHub activity, and bot groups from one place. The macOS app is free and open source, and runs bots locally using access to Codex, Claude Code, or OpenCode. The upcoming hosted plan charges a fixed price for the bot machine, while users bring their own AI provider key and pay that provider directly.

Sync your AI stuff away from your codebase

Dotlore is designed to keep AI files, folders, and custom patterns outside of your main codebase while ensuring they remain accessible. It syncs these elements across your devices using your preferred cloud service, such as iCloud or Google Drive. Currently, it prioritizes macOS with support for Windows and Linux planned for the future.

An IDE designed for agentic AI research.

AgentsDock is an integrated development environment (IDE) tailored for agentic AI research. It enables users to run Claude Code, Codex, and Cursor on their own machines, offering a unified workspace that can be accessed from various devices including desktop and mobile. This platform is especially beneficial for researchers who need to manage multiple AI agents and server connections seamlessly.

AI-powered automated testing platform that helps teams build reliable software faster.

AutoQA is an AI-powered automated testing platform that enables teams to streamline their quality assurance process. With features that allow users to create, run, and manage test plans through intelligent automation, AutoQA simplifies the testing of complex applications. The platform is suitable for project managers, QA teams, and software engineers who want to enhance their software delivery speed and reliability.

your agent works in your terminal, you keep the keyboard.

Conn is a terminal that you share with the agent you already use. It allows the agent to see what you see on screen and asks for permission before it types. You retain control with every keystroke, ensuring that you manage potentially risky commands while benefiting from the agent's assistance. It is designed for users of Claude Code, Codex, Cursor, and GitHub Copilot.

Free AI Excel assistant. Generate and explain Excel formulas, VBA macros and regex in seconds.

AIExcel is an online tool designed to help users generate and explain Excel formulas, VBA macros, and regex patterns quickly and easily. It caters to users of Microsoft Excel and Google Sheets, providing a straightforward interface that requires no sign-up or credit card. This platform stands out by offering a seamless experience without the usual barriers of account creation and usage limits, making it an excellent resource for anyone needing quick Excel assistance.

OASIS is an open-source, local-first agentic development workspace for coding, automation, and verif

OASIS is an open-source coding agent and development workspace that lets developers inspect repositories, plan and edit code, run approved tools, and verify changes. It can be used from the terminal or a local browser workspace, and supports multiple model providers and compatible endpoints. The runtime includes command approvals, file snapshots, checkpoints, verification gates, and safeguards for risky commands and weakened tests. It also supports MCP integrations, Agent Skills, hooks, bounded subagents, headless workflows, CI scripts, a GitHub Action, and a Node.js SDK.

Meet ERA. Your voice, your ideas, and a team of AI agents working together, right on your desktop.

ERA AI is a Windows desktop workspace that combines voice interaction, specialist AI agents, and desktop tools. It supports screen vision and OCR, video calls and object detection, and offers eight specialists focused on areas including coding, research, security, files, travel, shopping, careers, and the browser. Users can connect configurable AI providers with their own API keys and choose which tools to use. Desktop tools operate locally, while voice and AI features may connect to the configured provider; sign-in, licensing, and updates use the ERA backend. ERA is made for Windows 10 and 11.

Várias contas do Claude Code, revezando sozinhas: troca antes do limite, a sessão continua, logins c

Reveza AI automatiza a troca de contas do Claude Code antes que o limite seja atingido, garantindo que o trabalho não seja interrompido. Os usuários conectam suas contas uma vez, e a plataforma acompanha o uso, permitindo uma transição fluida entre contas. Os logins são mantidos seguros e cifrados localmente, e a experiência do usuário é enriquecida com um painel hospedado na nuvem para um gerenciamento eficaz das contas.

Capy is the best AI software engineer.

Capy is designed to orchestrate a fleet of cloud coding agents that assist teams in planning, building, and shipping software solutions. It integrates seamlessly with existing tools, enabling engineers to delegate tasks effectively while managing workflows efficiently. Capy helps in reducing busywork and enhances productivity for software development teams.

Set up Claude Code once. Your whole team gets it.

Trina AX puts Claude Code and Codex on a company server with shared files and team skills, while checking each person's folder permissions whenever the AI accesses files. Employees use it in a browser on desktop or phone, with a Windows installer available for those who want an app. It also supports other models and connected services.

Small computer jobs, done by an AI agent and checked before you get them.

Hire King AI offers fixed-price computer tasks completed by its autonomous AI agent. The listed jobs cover spreadsheet fixes, PDF table conversion, small Python scripts, and turning one public website into a clean dataset. Customers pay by card through Stripe, describe the task, and receive an email from King within 24 hours to confirm details and collect files. Each job has a defined scope, delivery timeframe, and one included revision. King checks the work before delivery, and results are sent by email. The service does not take tasks involving scraping behind logins or against site terms, password cracking, graded schoolwork, spam, fake reviews, or illegal activity.

可持久化任务栈

This site is a repository and marketplace for the @ray1270/dsh-task-stack plugin, designed for use with the DeepSeek Harness Agent. It allows users to manage multi-step tasks within a session, with features like task focus management and state recovery across sessions. The site serves developers interested in enhancing their agent workflows with structured task management solutions.

Combine LLMs and code to create simple automations

Magic Loops is a no-code platform that enables users to build AI-native applications quickly and efficiently. Unlike traditional coding methods, it allows anyone to create professional applications in minutes. The platform is designed for a wide range of users, from hobbyists to professionals, enabling them to develop various apps without writing any code.