Projects sharing features with Computer

rlm is an LLM code execution engine and orchestration framework designed to coordinate multiple language model calls and recursive sub-tasks through a programmable environment. It provides a sandboxed REPL environment and a recursive context processor to handle inputs that exceed standard token limits by programmatically decomposing prompts. The project differentiates itself through a reinforcement learning training harness used to teach models how to utilize recursive calls and code execution. It includes a reasoning visualization system that records and renders execution trajectories to ana

microsandbox is a platform that runs untrusted code inside hardware-isolated microVMs, each with its own kernel, filesystem, and network stack. It boots directly from standard OCI container images, supports copy-on-write filesystem layers, and integrates with AI agents to execute tool calls and generated code in isolated environments with secret protection. What sets microsandbox apart is its host-side network proxy that enforces firewall rules, intercepts DNS, inspects TLS traffic, and injects secrets at the network boundary without exposing them inside the VM. It provides SSH access to micr

Acontext is an LLM orchestration backend and agent memory framework designed to manage session state and knowledge for AI agents. It functions as a context manager and orchestration layer that integrates model providers with a secure code sandbox and a zero-knowledge data store. The project is distinguished by its approach to knowledge distillation, capturing agent learnings as reusable Markdown skills and structured memory files. It provides a secure execution environment where shell commands and scripts run in isolated containers with the ability to mount these persistent skill files direct

ms-agent is an LLM agent framework and multi-agent orchestration system designed to build autonomous entities that combine large language models with tool calling and structured workflows. It serves as a tool integration platform and workflow engine for executing complex tasks through the coordination of specialized agents. The project distinguishes itself through a multimodal agent workflow engine capable of automating the production of text, images, and video. It features a sandboxed code execution environment for running generated code and quantitative data analysis in isolated containers,

Chat on Steroids is a desktop companion app that connects ChatGPT to the user's own machine through local Model Context Protocol servers. It lets the chat read, search and patch files in a real project, run tests, keep terminals open, and observe or control the desktop, with a companion Chrome extension that drives real browser tabs in the background without changing the user's selected tab or foreground window. Access is split across separate Core, Desktop and Plugins connections, each authorized independently. What sets it apart is the permission layer around that access. File access is res

OpenJarvis is a local-first personal AI assistant and agent framework. It runs language models on a user's own laptop, workstation, or server, keeps agent memory and state on the device, and treats cloud services as optional extras. Agents reason over multiple steps, call tools, search documents, and can run on demand, on a cron schedule, or continuously in the background. Core functionality stays available offline. A unified inference layer connects local runtimes and optional cloud providers behind one interface. It detects the installed GPU vendor, model, and memory, then recommends a suit

mini-swe-agent is an autonomous software engineering system designed to develop features and fix bugs by combining large language models with a bash interface. It operates as an agentic framework that executes coding tasks and documentation updates through a continuous cycle of model reasoning and tool execution. The project differentiates itself with a strong focus on safety and evaluation, utilizing container-based sandbox execution via Docker or Singularity to isolate command execution. It includes a batch-parallel evaluation harness to measure code-fixing accuracy against standardized sof

Judge0 is an online code execution engine and multi-language compiler API designed to compile and run source code within isolated sandboxes. It functions as an asynchronous job processor that handles code submissions via a queue and provides a secure environment to run arbitrary programs while preventing unauthorized system access. The system distinguishes itself through a multi-stage compilation pipeline and a flexible execution model that supports both single-file submissions and multi-file program execution via archives. It employs an isolate-based sandboxing mechanism to enforce strict ha

s3fs-fuse is a FUSE-based filesystem that mounts Amazon S3 buckets and S3-compatible object storage as local directories. It functions as a cloud object storage gateway, translating standard POSIX filesystem calls into RESTful API requests to allow remote objects to be accessed as if they were local files. The project enables the mapping of remote object stores to local paths on Linux, macOS, and FreeBSD. It supports integration with third-party S3-compatible providers through the use of custom service URLs and specialized request styles. The system includes capabilities for local data and m

This project provides a secure, containerized execution engine designed to run untrusted code within isolated environments. It functions as a library for integrating code interpretation into autonomous agents and intelligent assistant workflows, ensuring that host systems remain protected while enabling dynamic data processing and file manipulation. The platform distinguishes itself through a multi-backend architecture that abstracts diverse container runtimes, allowing for flexible deployment and automated backend failover. It supports interactive, multi-turn workflows by maintaining persist

mistral.rs is an inference engine for large language models that runs locally and exposes models behind OpenAI and Anthropic-compatible APIs. It serves as a multi-model serving platform, capable of loading several models in a single server process with per-request routing and on-demand loading and unloading. The engine supports multimodal inference, processing text alongside images, video, audio, and speech inputs, and includes a quantized model deployment runtime that reduces memory use and speeds up inference on consumer hardware. The project distinguishes itself through an agentic tool exe

TabFS is a browser automation tool and virtual filesystem that maps browser tabs and page elements to files. It provides a FUSE-based bridge that allows external scripts and tools to interact with a browser's JavaScript runtime through standard file system operations. The system enables the execution of JavaScript expressions and the manipulation of DOM elements by reading and writing synthetic files. Users can control tab state, create new tabs, and inspect metadata such as URLs and titles using a file manager or shell. The project covers capabilities for browser automation, including acces

x-cmd is an AI agent orchestrator, cloud infrastructure CLI, and cross-platform package manager that provides an enhanced POSIX shell toolkit. It integrates large language models directly into the terminal for chatting, code generation, and the execution of agentic workflows, while offering a framework for building interactive terminal user interface components. The project distinguishes itself by deploying containerized AI agents within isolated sandboxes, provisioning them with specialized skills and headless browser automation capabilities. It further streamlines development through a unif

This project is an interactive programming curriculum and educational system designed to teach computer science and software engineering. It provides a structured set of courses and professional roadmaps focused on backend engineering, DevOps, and systems fundamentals. The platform is distinguished by an AI-powered coding tutor that provides Socratic guidance and contextual hints to help students find solutions independently. It features a browser-based code sandbox using WebAssembly to eliminate local environment setup, alongside automated test-based grading and spaced-repetition logic to re

Upspin is a decentralized naming and storage system that provides an end-to-end encrypted file system. It assigns every user a unique identity and organizes files within a global, permissioned namespace where data is encrypted on the client before transmission. The system separates identity resolution from data storage using a public-key identity provider and a key-server architecture. This allows for decentralized identity management and the resolution of usernames to specific directory and storage server addresses. The project includes a hierarchical access control system that manages read

Letta is a framework for building, deploying, and managing autonomous AI agents that maintain persistent state across long-term interactions. It provides a comprehensive suite of primitives for defining agents with configurable personas, modular memory blocks, and tool-use capabilities, enabling them to retain user preferences and conversation history over extended sessions. The platform distinguishes itself through its advanced memory management and orchestration capabilities. It allows agents to autonomously update their own memory, perform retrieval-augmented generation, and coordinate com

Agent-native is a TypeScript framework for building applications where AI agents and the user interface share the same actions, state, permissions, and data. Each capability is defined once as an action. Agents call it as a tool, and application code calls it as a typed function, so both work from one source of truth with the same validation and logic. The same action definitions can be served over HTTP, MCP, A2A, and the command line, which lets compatible agents and clients call them without the logic being rewritten. Work can be delegated to specialist agents that run in parallel, either i

Goofys is a POSIX-compliant cloud object storage gateway that presents remote storage buckets as local system directories. It implements a user-space filesystem that maps S3-compliant and Azure Blob storage services to local mount points, allowing remote objects to be accessed through standard file system operations. The project provides specific mounting capabilities for Amazon S3, Azure Blob Storage, and Azure Data Lake accounts. It utilizes a FUSE-based implementation to interface cloud object storage with the operating system kernel. The system includes performance optimizations such as

This project is a comprehensive framework for building and managing autonomous agent systems. It provides a unified architecture for orchestrating multi-agent societies, where specialized agents collaborate through roleplay to decompose and solve complex tasks. The system integrates language models with external environments, enabling agents to perform real-world actions through a standardized tool-calling abstraction layer. The framework distinguishes itself through its focus on iterative reasoning and data reliability. It employs automated feedback loops to refine agent outputs and self-eva

Capnweb is a distributed object communication library and Cap'n Proto RPC framework. It enables type-safe remote procedure calls between clients and servers using shared schemas and generated stubs to invoke methods on remote objects as if they were local. The project utilizes an object-capability security model to govern access to remote resources through unforgeable tokens. It provides a bidirectional network layer that multiplexes asynchronous calls and data streams over persistent WebSocket connections and includes a remote resource lifecycle manager that uses reference counting to automa

The Gemini Cookbook is a comprehensive collection of implementation patterns, code samples, and development guides designed for building applications with Google Gemini models. It serves as a central resource for developers to integrate multimodal generative artificial intelligence into their software, providing the necessary frameworks to manage model interactions, stateful workflows, and structured data extraction. The repository distinguishes itself by offering specialized toolkits for autonomous agent orchestration, enabling the construction of agents that can execute code, browse the web

Afero is a Go library that provides a unified filesystem abstraction, allowing applications to interact with local disk, in-memory storage, cloud services, archives, and remote systems through a single, consistent interface. At its core, it defines a standard interface that all filesystem backends implement, enabling developers to swap storage implementations without changing application code. The library distinguishes itself through its composable architecture, which includes layered filesystem composition for creating cached, sandboxed, or restricted storage views. It offers a copy-on-write

The AWS Cloud Development Kit is an infrastructure-as-code framework that enables developers to define and provision cloud resources using familiar programming languages. By utilizing construct-based synthesis, it translates high-level, object-oriented code into declarative templates, allowing for the automated management of complex cloud environments through a centralized, code-driven control plane. The framework distinguishes itself through its ability to model infrastructure as a dependency-aware resource graph, ensuring that components are provisioned and updated in the correct order. It

This repository is a long-form technical book about how a production AI coding agent runtime, often called an agent harness, is built. It dissects a real-world coding agent's architecture chapter by chapter and turns it into transferable lessons, and it also works as a step-by-step guide to designing and building your own harness around a language model. Python scripts in the repository support the book itself rather than the agent. The book's core subject is the machinery around the model: a streaming conversation loop that repeats tool calls until the model returns plain text, a tool protoc

Dynamiq is an agent development platform designed for building, orchestrating, and monitoring autonomous agents. It provides a framework for constructing complex, multi-step workflows using a graph-based engine that supports conditional branching, feedback loops, and iterative task execution. The platform distinguishes itself through its focus on secure, private infrastructure, allowing for the deployment of language models and orchestration services within virtual private clouds to maintain data sovereignty. It integrates retrieval-augmented generation pipelines to ground model responses in

Midscene is a multimodal automation framework designed to enable AI agents to perceive, navigate, and manipulate graphical user interfaces across web, mobile, and desktop environments. By leveraging vision-capable AI models, the platform interprets interface screenshots to execute tasks based on natural language instructions, removing the reliance on traditional, brittle code-based selectors. The framework distinguishes itself through its ability to decompose high-level goals into autonomous, multi-step sequences that function consistently across diverse platforms. It provides a visual ground