Explore the directory

A NetBeans IDE plugin that adds an AI pair-programming chat interface inside the editor, connecting through the Agent Client Protocol to coding harnesses such as Claude, Cursor, OpenCode, Gemini and Goose.

• Coding agents
• Editors

A command-line client for the Harmont CI platform that runs pipelines defined in Python or TypeScript, either locally in Docker or on Harmont Cloud's managed runners, with DAG-based parallelism and layer caching.

• CI/CD

A set of UI components for editing, viewing and testing agents that implement the Agent2Agent (A2A) protocol, usable as an npm package or a ready-to-run Docker image with a live playground.

• Agent frameworks
• UI kits

A Rust library that grafts textures from another GPU API, OpenGL, Vulkan, Metal, or DirectX 12, onto host-owned wgpu textures, letting a wgpu-based Rust GUI embed a renderer such as Servo without a CPU copy.

• Libraries

A Python library and CLI that runs mixture-of-experts language models larger than a consumer GPU's memory by offloading unused experts to CPU, reaching real-time-usable throughput on an 8 GB laptop GPU.

• Inference

A standalone Python system that keeps a vector database automatically and incrementally synced with a set of source files, watching for changes and re-embedding only what changed.

• RAG
• Databases

A static analysis and attack-surface scanner for AI agents that maps an agent's tool access, prompt risks, and MCP integrations from source code, then turns material authority changes into CI/CD pass, review, or block decisions.

• Security
• CI/CD

A local-first, open-source visual canvas for orchestrating teams of real Claude Code agent processes, letting a user drag agents, skills, and safety rules onto a board and run a governed multi-agent workflow.

• Agent frameworks
• Workflow automation

A tool that turns a local folder into a public Model Context Protocol endpoint over a tunnel, exposing five deliberately minimal tools for running shell commands and transferring files, as a VS Code extension or standalone desktop app.

• MCP servers
• Networking

A self-hosted, privacy-focused retrieval-augmented assistant, written in C++, that indexes a codebase locally and answers questions about it using local or cloud language models instead of a paid, cloud-only coding assistant subscription.

• Agent memory
• RAG

A local-first retrieval-augmented research assistant that helps scientists and students search, understand, and organize academic literature across sources such as Semantic Scholar, PubMed, and arXiv, exposed through an MCP server and REST API.

• MCP servers
• RAG

A Python multi-agent orchestration framework that turns a YAML-defined agent team into a runnable command-line agent or an A2A server, with sub-agent delegation, persistent per-project memory, and many supported model providers.

• Agent frameworks

A self-hosted server that runs one AI agent and lets it work in a server workspace or on paired Linux, macOS, and Windows computers, reachable through a browser, Discord, or DingTalk, using an Anthropic-compatible API.

• Agent frameworks
• Containers

A JetBrains IDE plugin that integrates the OpenCode AI coding assistant, managing the OpenCode server's lifecycle and showing its session changes as inline diffs and a three-panel diff view inside the editor.

• Coding agents
• Editors

A desktop AI agent framework with one-click installation, visual drag-and-drop workflow building, markdown-based skill extensions, a personal knowledge base, and multi-channel chat across WeChat, Slack, Discord, and Telegram.

• Agent frameworks
• Workflow automation

A local AI agent terminal built on LangGraph that organizes reasoning and tool calls as a state machine, restricting tool execution to a sandbox and interrupting the agent when it detects a repeating loop.

• Coding agents

A message broker built specifically for multi-agent systems that unifies agent registration, capability-based discovery, and reliable asynchronous messaging, with native support for the A2A agent-to-agent protocol.

• Agent frameworks
• Backend

A workspace for running and benchmarking local LLMs on Apple Silicon with MLX, built to measure which local model can handle routine coding operations well enough to replace paid cloud requests in an internal coding tool.

• Inference
• AI evals

An Unraid Community Applications template and Docker image that bundles Mem0's OpenMemory memory layer with its Qdrant vector store, API, and dashboard into one self-hosted container for homelab use.

• Agent memory
• Containers

An out-of-the-box fine-tuning framework, built on Hugging Face Transformers, PEFT, and DeepSpeed, for training large language models to perform binary classification on medical text such as disease-risk prediction.

• Models

A Codex and Claude Code skills suite that turns MATLAB and Simulink into an agent-operable engineering workbench, packaging modeling, code generation, and validation workflows so every task ends with a runnable command and saved evidence.

• Agent skills

A lightweight Tauri desktop control panel for running llama.cpp locally, letting a user discover GGUF models, configure and preview the launch command for llama-server, and save per-model presets.

• Inference

A self-hosted, read-only AI agent for Kubernetes that answers questions using a local Mistral model through Ollama and reasons about a cluster's actual live pods, events, and logs rather than generic advice.

• Containers
• Monitoring

A set of Claude Code skills that turn a Lanhu design-tool page link into Vue or React frontend code, reusing existing project components and mapping raw design values to design tokens instead of hard-coding them.

• Agent skills