Archestra is an open‑source, enterprise‑grade AI platform that bundles a chat UI, LLM gateway, agent runtime, RAG knowledge base, mini‑app builder, and full security/observability stack. It lets companies expose LLMs to users and internal tools behind a single, SSO‑protected URL, with cost limits, sandboxed execution, and deterministic guardrails.

A proxy service that converts ChatGPT Web backend capabilities into an OpenAI-compatible API, supporting chat, images, audio, and file-based Q&A.

A design skill for AI coding agents that prevents generic AI-generated UIs by using computational palette generation, contrast verification, and persistent design memory.

A secure Telegram client for the OpenCode CLI that allows users to remotely run AI coding tasks, manage sessions, and monitor progress from their phones.

A browser extension that suggests similar GitHub repositories using a vector database and a model trained on 300 million stars.

An open-source AIOps and alert management platform that centralizes, deduplicates, and uses AI to correlate and summarize alerts from multiple observability tools.

An open-source calendar syncing tool that aggregates events from multiple providers and pushes them to destination calendars to prevent scheduling conflicts.

A PyTorch native platform for rapid experimentation and large-scale training of generative AI models, providing a clean-room implementation of native scaling techniques.

A FastAPI-based API server that provides an OpenAI-compatible interface for generating text and embeddings using the ExllamaV3 backend.

An open-source platform for organizations to manage, secure, and govern their AI ecosystems, providing centralized control over LLM access, MCP servers, and AI activity auditing.

An intelligent routing layer for LLM inference that optimizes request placement based on load and prefix-cache locality to improve serving performance.

A headless CLI client for the Agent Client Protocol (ACP) that provides a standardized interface for managing sessions and interacting with multiple AI coding agents.

A Claude Code plugin that implements an architect pattern, orchestrating tasks between Fable 5.1 and GPT-5.6 models to optimize cost and quality through cross-vendor review.

An official Rust SDK for the Model Context Protocol (MCP) that enables developers to build standardized servers and clients for sharing tools, resources, and prompts with AI models.

A security-first MCP server that provides LLM agents with controlled SSH access to remote hosts through command classification, policy-based authorization, and human-in-the-loop approval.

A free and open-source framework for streamlining the post-training and fine-tuning of large language models and multimodal models using a configuration-driven approach.

MLC LLM is a universal deployment engine and ML compiler that enables high-performance native execution of large language models across diverse hardware and platforms.

An Expert Parallelism communication library for MoE models that eliminates load imbalance via dynamic redundant experts and zero-copy communication.

An AI-powered analysis tool for Android Perfetto traces that uses natural-language queries to identify performance bottlenecks like jank and ANRs with evidence-backed reasoning.

An adapter that connects the pi coding agent to ACP-compatible clients like the Zed editor, enabling IDE-integrated AI coding assistance.

A tool that allows developers to copy UI elements from a browser and provide AI agents with the exact source code location and component stack.

An open-source Python library for testing and evaluating agentic systems, providing tools for automated red-teaming, vulnerability scanning, and LLM-as-judge evaluations.

A design engineering system for AI coding agents that provides design rules and verification tools to ensure AI-generated UIs are professional and production-ready.

A MySQL-compatible state store for AI agents featuring hybrid vector and full-text search, high-concurrency streaming writes, and copy-on-write sandboxes for safe state exploration.

A benchmark for measuring whether AI models can detect and reject nonsense premises in questions instead of confidently accepting invalid assumptions.

A comprehensive suite of custom nodes for ComfyUI that adds professional image editing, 3D scene building, and workflow management tools directly into the AI generation process.

An MCP server that enables AI agents to build, modify, and manage Power BI semantic models using natural language and automated workflows.

A high-performance request forwarding system for large-scale vLLM deployments that provides advanced load balancing and prefill-decode disaggregation support.

A lightweight extension for PyMuPDF that converts PDFs and other documents into structured Markdown, JSON, and plain text optimized for RAG pipelines and LLM ingestion.

PegaInfer is a Rust‑based, CUDA‑accelerated LLM inference server that provides an OpenAI‑compatible API, supports multiple model families (Qwen3, Gemma 4, DeepSeek‑V2‑Lite, etc.), and offers production‑grade performance without any Python runtime.

A production-grade Kubernetes skill for Claude Code and Codex that prevents AI hallucinations and ensures secure, reliable manifests through a failure-mode-first workflow.

A Claude Code plugin that routes bulk execution and token-heavy tasks to the Antigravity CLI (Gemini), allowing Claude to act as a conductor for verification and judgment while reducing costs.

A collection of SQL and AI-based models for detecting financial risk, fraud analytics, and regulatory compliance in banking and lending.

A local-first MCP server and CLI that provides AI agents with 106 deterministic astrology and metaphysics calculation tools to prevent hallucinations and ensure traceable results.

A private, open-source desktop app for running local LLMs with support for multiple backends, multimodal input, tool-calling, and local fine-tuning.

An open-source API layer that turns local LLM inference servers into production-ready AI applications by providing RAG, tool use, and Claude API compatibility.

A premium desktop workspace and GUI for OpenCode that transforms the AI coding terminal tool into a full-featured integrated development environment.

A skill layer for AI agents that provides a registry and resolver to find, audit, and install the most suitable and trusted reusable skills for specific tasks.

herdrm is a native macOS SwiftUI app that serves as a graphical console for the herdr AI‑coding‑agent runtime. It lets you attach to real PTYs of agents (Claude, Codex, Gemini, etc.) on the local Mac or any SSH‑accessible machine, provides a sidebar showing devices, spaces, agents and terminals, supports file transfer, paste‑of‑files/images, notifications, and global search. Install via Homebrew or a signed zip, requires macOS 14+ and a running herdr daemon. Early‑stage but functional, licensed for non‑commercial use.

A Python library for accelerator-oriented array computation and program transformation, designed for high-performance numerical computing and large-scale machine learning.

A high-performance RLHF framework using Ray and vLLM to scale reinforcement learning for LLMs and VLMs, featuring a unified agent-based design for single and multi-turn training.

An operating system for AI agents that provides a daemon-based runtime to manage persistent sessions, automation, and memory for ACP-compatible agent CLIs.

The official desktop application for ComfyUI that provides a self-contained, GPU-ready environment to install, run, and manage multiple independent AI generative workflows without dependency conflicts.

A local CLI tool that indexes and searches past coding agent sessions to provide agents and developers with exact recall of historical decisions and tool calls.

An infrastructure for agentic analog and mixed-signal design that allows LLM agents to drive Cadence Virtuoso for automated schematic, layout, and simulation flows.

BAML is a programming language designed for AI agents that provides a strong type system and built-in evaluation tools to ensure reliable, structured outputs.

A scalable Kubernetes scheduler that optimizes GPU resource allocation and fairness for large-scale AI and machine learning workloads.

A collection of repair scripts and agent skills for Windows Codex Desktop that fixes broken features like Fast Mode, Computer Use, and plugin markets after software updates.