A portable AI coding workspace that bundles the official Claude Code runtime and Agent SDK, allowing it to run from a USB drive with support for multiple AI providers.

ccteam is an orchestration layer that turns separate AI coding CLIs into a coordinated team, allowing agents to operate across different vendors and machines from a unified console.

NeMo Guardrails is an open‑source Python toolkit from NVIDIA that lets developers add programmable safety and control rules (guardrails) to LLM‑based chat applications. It supports input, dialog, retrieval, execution, and output rails, uses a simple DSL called Colang for defining flows, works with many LLM providers, offers a CLI/server with an OpenAI‑compatible API, and includes evaluation tools. The library is aimed at building trustworthy, secure conversational AI.

A bridge that allows you to remotely control AI programming CLIs from Lark/Feishu, streaming terminal output as interactive cards and providing a web terminal for mobile and desktop access.

A model-independent safety control plane for AI agents that enforces deterministic policies and cryptographic human approval outside the language model.

A Kubernetes-native framework for building, deploying, and managing AI agents as declarative custom resources.

An open-source platform for AI production that allows companies to trace, test, route, and govern all LLM calls and coding agents.

An open-source dependency manager for AI agents that allows developers to declare and share agent configurations, skills, and plugins via a portable manifest file.

An open-source anti-detect browser launcher using a patched Chromium engine to provide isolated, spoofed browser identities for web scraping and multi-accounting.

A microVM sandbox runtime for AI agent fan-out that uses snapshot copy-on-write to spawn hundreds of isolated, warmed runtimes in milliseconds.

An open-source git extension that provides line-level attribution for AI-generated code, linking every line to the agent, model, and prompt that created it.

A tool that detects which LLM is powering an API by comparing its response distributions to known model benchmarks.

A tool that enables DLSS 5 neural rendering in games without native DLSS support by bridging ReShade data to DLSS calls.

A CLI and TUI switchboard for managing API providers, accounts, and system prompts across multiple AI coding assistants like Claude Code, Codex, and Gemini.

A multi-engine AI coding desktop client that brings command-line AI coding runtimes like Claude Code and DeepSeek Harness into a single graphical interface.

A Python framework that transforms scripts into interactive web apps, enabling data scientists to build and share dashboards and AI chat apps without front-end experience.

A multimodal AI lakehouse and vector database built on the Lance columnar format for fast, scalable search over petabytes of vectors and multimodal data.

A reference implementation for production vLLM deployment on Kubernetes, providing a scalable inference stack with request routing and observability.

A local AI application and inference engine that lets users run open-weight LLMs privately on their own devices while providing an OpenAI-compatible API for other agents.

A local-first, self-hosted workflow engine that turns existing scripts and containers into production DAGs using declarative YAML, requiring no external database.

A visualizer for Claude Code and Codex sessions that transforms text transcripts into interactive, live flow graphs for monitoring and reviewing agentic workflows.

A deep learning library built on PyTorch that provides high-level components for rapid productivity and low-level components for research flexibility.

A Markdown-native task manager and Kanban visualizer that structures AI agent workflows around spec and plan reviews to prevent code bloat and review fatigue.

An open-source benchmark for evaluating AI browser agents on everyday online tasks across live websites, featuring a 5-layer recording pipeline and agentic evaluation.

An algorithmic conversation compactor for the Pi AI agent that reduces token usage and prevents history loss using deterministic extraction instead of LLM summarization.

A library of structured engineering skills for AI agents that replaces generic AI outputs with disciplined habits like systematic debugging and design pressure-testing.

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.

ExploitGym is a large-scale benchmark of real-world vulnerabilities used to evaluate the ability of AI agents to develop functional security exploits.

Zvec is an open‑source, in‑process vector database from Alibaba. It stores dense or sparse embeddings, offers lightning‑fast ANN and full‑text search, supports hybrid queries, and provides SDKs for Python, Node.js, Go, Rust, and Dart. The library runs locally on Linux, macOS, Windows, Android, and iOS, with durable WAL storage and concurrent read access. Ideal for LLM‑based retrieval, edge devices, or any app needing fast, embedded similarity search.

A collection of 14 expert skills and an enforcement layer for Claude AI, enabling it to build, validate, and deploy production-ready n8n workflows via the n8n-mcp server.

An open-source, AI-integrated terminal that combines a command-line interface with context-aware AI assistance, durable SSH sessions, and built-in graphical editors.

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

An all-in-one Android automation and control platform that integrates UI automation, remote desktop, packet capture, and AI agents into a single unified service.

A Python SDK that provides a single API gateway to access diverse data domains including legal cases, enterprise records, and academic research, specifically optimized for AI agents.

A local-first screen capture tool for macOS and Windows that integrates screenshots, annotations, OCR, and region recording.

An open-source RL framework that uses GRPO to train multi-step AI agents to learn from experience and improve reliability in real-world tasks.

A tool that converts software repositories, pull requests, and task seeds into executable, verifiable RL environments in the Harbor format for training and evaluating AI agents.

A comprehensive framework and codebase for developing ML-driven trading strategies, covering the entire pipeline from data acquisition and signal research to live production deployment.

A lightning-fast search engine that provides full-text, semantic, and hybrid search capabilities with search-as-you-type performance and AI-ready integrations.

A skill for AI coding agents that scaffolds a Next.js editor to create and export production-ready marketing screenshots for the Apple App Store and Google Play Store.

An open-source static analysis CLI that detects dead code, security flaws, and AI-generated code hallucinations across multiple languages to ensure code quality before merge.

A high-performance file and content search toolkit that provides typo-resistant, frecency-ranked search for humans and AI agents.

A local-first session browser and cost tracker for AI coding agents, providing unified search, analytics, and spend monitoring across multiple tools.

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

An open-source AI orchestration framework for building production-ready RAG systems, autonomous agents, and LLM applications in Python.

A rigorous workflow assistant for math-modeling contests that uses a gated pipeline and auditing layers to ensure data consistency and reproducibility while keeping modeling decisions in human hands.

A multi-tier framework for evaluating AI agent skills to ensure they are safe, non-redundant, and effectively improve agent performance through validation, deduplication, and live sandboxed testing.

An AI-powered reverse-engineering agent that uses Ghidra and LLMs to reconstruct and validate C/C++ functions from compiled binaries through an autonomous pipeline.