A desktop application for managing multiple OpenAI Codex accounts, allowing users to easily switch between them and monitor usage quotas.

An LLM inference framework for CPU servers that optimizes long-horizon inference by trading storage for computation to outperform GPUs in long-context scenarios.

A test runner for the Agent Skills ecosystem that empirically measures the performance lift of domain knowledge files by comparing model outputs with and without those skills.

OpenProgram is a Python framework that turns ordinary functions into LLM‑backed agents. The docstring becomes the system prompt, `llm()` calls the model, and the surrounding logic stays in deterministic code. It provides four core mechanisms – Agentic Functions, a DAG‑based context graph, workflow gates that enforce validation, and a process‑wide event bus – enabling multi‑agent, self‑modifying, and proactive AI assistants. Installable via a one‑line script, it ships with a terminal and web UI, supports Anthropic/OpenAI/Gemini providers, and includes plug‑in harnesses for GUI automation, research workflows, and memory. Licensed AGPL‑3.0 and accompanied by a KDD 2026 workshop paper.

A model-neutral coding runtime using the Model Context Protocol that gives AI agents secure access to read, patch, and execute commands within a codebase.

A local-first AI subtitle generator that provides on-device transcription and translation with direct integrations for DaVinci Resolve and Adobe Premiere Pro.

A lightweight OS-level sandboxing tool that enforces filesystem and network restrictions on processes to enable safer AI agents and MCP servers.

A system-level AI Agent for Android that allows users to bring their own LLM to control device APIs, automate GUIs, and run a full Linux environment.

A tool for managing a pool of reusable, isolated git worktrees, providing AI agents and developers with instant, cache-preserving environments without cloning conflicts.

An open-source front-end quality system that turns best practices into a practical review workflow for humans and AI agents via a website and MCP server.

Ray is a unified framework for scaling Python and AI applications from a single laptop to a distributed cluster, providing a core runtime and specialized libraries for ML compute.

An all-in-one training suite for diffusion models that allows users to fine-tune image, video, and audio models on consumer-grade hardware via a GUI or CLI.

An open-source Kubernetes UI that provides real-time cluster visualization, resource management, and GitOps monitoring as a single binary with no required cluster-side agents.

A programmable TypeScript harness for building autonomous agents with built-in support for secure sandboxes, durable sessions, and tool integration.

GoRaven is a self-hosted AI agent workstation for teams that enables agents to execute real-world tasks like coding, running commands, and calling APIs instead of just chatting.

A ready-to-go translation and OCR tool developed with WPF for easy text translation and image text recognition.

A collection of composable skills for coding agents that improve their autonomy, visual planning, and efficiency in software development tasks.

An open-source platform for securely running and managing Model Context Protocol (MCP) servers using container isolation and centralized governance.

A provider-neutral framework of best practices and blueprints for designing production-safe agentic harnesses that separate model proposals from runtime execution and authorization.

An open-source desktop application for building and deploying a multi-agent AI workforce to automate complex digital workflows.

An MCP server that enables AI assistants to search, edit files, and execute terminal commands on your local machine, featuring advanced support for Office documents and process management.

A tool that converts YouTube, Bilibili, and local videos into illustrated Markdown or HTML notes using speech recognition.

A native desktop installer and manager for AI agent personas that synchronizes specialist personas across various AI coding tools like Cursor and Claude Code.

A lean, high-performance terminal coding agent harness written in Go that provides a fast, low-overhead environment for LLMs to edit code and execute commands.

A general-purpose problem-solving engine that uses a state-space search and blackboard architecture to autonomously find paths to goals, specifically validated for AI penetration testing.

A distilled cybersecurity knowledge base of 101 attack playbooks and methodologies designed to be loaded as skills by AI agents for penetration testing and security research.

An autonomous multi-agent system that conducts scientific research by crawling literature, running experiments, and writing publication-grade LaTeX reports.

A powerful OSINT suite for deep email and username intelligence that maps digital footprints across 465+ platforms with automated pivoting and AI agent integration.

A GGML/C++ implementation of NVIDIA's Kimodo model that generates character and robot motion animations from text prompts.

An MCP server that enables AI agents to natively control the Windows OS, perform UI automation, and manage files without needing computer vision.

An agent-project workbench for DeepSeek Harness that provides a dockable multi-window workspace and a real-time control room for monitoring AI projects.

JIT-Agent is a meta-agent that dynamically generates and evolves task-specific executable harnesses to optimize the performance of agentic LLMs.

An open-source coding agent for the terminal that can read repositories, edit files, and run commands to automate development tasks.

A collection of CLI harnesses that enable AI agents to control WPS/Microsoft Office, Zotero, Adobe Illustrator, and Adobe Photoshop via Windows COM automation.

An open-source PostgreSQL extension that enables vector similarity search, allowing users to store embeddings and perform nearest neighbor queries within their relational database.

An all-in-one video subtitle tool that uses LLMs for speech recognition, subtitle optimization, and translation to automate the video captioning process.

A reactive Python notebook that ensures consistency and reproducibility by automatically running dependent cells and storing notebooks as pure Python files.

A native macOS application that uses local AI models to provide instant, privacy-focused voice-to-text transcription and a conversational AI assistant.

An open-source library for intelligent LLM routing that dynamically selects the most suitable model for each query to optimize inference cost and performance.

A browser bridge and extension that connects DeepSeek Harness to Chrome or Firefox, allowing AI agents to operate within the user's active, logged-in browser sessions using a text-only interface.

A local MCP server that connects ChatGPT to your local machine and allowed repositories, enabling the AI to read, search, and edit code directly.

A gateway that converts the official ChatGPT web interface capabilities into an OpenAI-compatible API, featuring a management console for multi-account and proxy orchestration.

FlashInfer is a Python library that supplies ultra‑fast, hardware‑aware CUDA kernels for LLM inference—covering attention, GEMM, MoE, sampling, and related ops. It automatically selects the best backend (FlashAttention‑2/3, cuDNN, CUTLASS, TensorRT‑LLM), supports low‑precision FP8/FP4, works with all modern NVIDIA GPUs, and is already used in major serving frameworks like vLLM and TensorRT‑LLM.

A Herdr plugin that allows users to annotate terminal text and review AI agent replies, sending feedback directly back to the agent.

A private, offline AI assistant for iOS and Android that runs language models and text-to-speech entirely on-device.

An inference library for running local LLMs on consumer GPUs featuring EXL3 quantization, parallel inference, and CPU offloading for large MoE models.

An LLM application development framework for Golang that provides tools for building AI agents, multi-agent coordination, and graph-based workflows.

A kinematic motion diffusion model that generates controllable 3D human and robot motions using text prompts and kinematic constraints.