Explore the directory

A CUDA and C++ implementation of quantized large language model training, aimed at single-node, multi-GPU training setups, inspired by llm.c.

• Models

An Odoo module that automates database backups to local storage or a remote FTP server, with versions maintained for each supported Odoo release.

• Databases

A shared configuration bundle for Claude Code and Codex containing global instructions, coding rules, plugins, skills, status lines and correction memory, installed by asking the user's own agent to read the install instructions and offer a menu of options.

• Agent skills

An open-source desktop client for connecting to and exploring vector databases, running vector and hybrid searches, and visualizing embeddings in 2D or 3D.

• Databases

A public-preview toolkit that builds a Hugging Face model into a versioned TensorRT inference bundle in two commands, with C++ task APIs for text generation, transcription, image and video generation, segmentation, embedding and forecasting.

• Inference

A self-hosted AI gateway and application backend that an AI agent can operate end to end through MCP: routing model traffic, defining data models that become database tables and CRUD APIs, and assembling React frontend blocks.

• LLM gateways
• Backend

An OpenCode plugin that keeps a task and its unfinished work connected across turns, using a Commander role for the overall goal and optional Planner, Worker and Reviewer roles when a task is complex enough to benefit.

• Agent skills

An open-source, local-first personal AI assistant aimed at long-term use, that keeps conversations, files, research and generated documents together in one task and can follow up later through scheduled work.

• Agent frameworks

A self-hosted geospatial data catalog that brings files, database tables and remote service snapshots into one searchable index, with a browser-based map builder and OGC and STAC APIs for existing GIS tools.

• APIs
• Data engineering

A hands-on quick-start guide for using the DeepSeek API, covering environment setup, an OpenAI-compatible client, and optional local GPU deployment with CUDA, Miniconda and Jupyter Lab.

• Tutorials

A self-hosted platform combining AI agents with a data stack: multi-agent workflows and RAG knowledge bases on one side, and an ETL pipeline, a lakehouse, BI dashboards and no-code machine learning on the other, governed by one shared catalog and audit trail.

• Agent frameworks
• Data engineering

An evolutionary runtime that breeds and grades populations of autonomous coding agents against a benchmark over repeated generations, evolving each agent's system prompt, code, tools and dependencies rather than just its output.

• Agent frameworks
• AI evals

A CLI scaffold that adds persistent memory, identity and coordination to AI coding agents built on Claude Code, so multiple agents share one project filesystem and can pick up work another agent left off.

• Coding agents
• Agent memory

A suite of NuGet packages for F# that generates types from a database schema and provides strongly typed, compile-time-checked query expressions for SQL Server, PostgreSQL, SQLite, Oracle, or MySQL.

• Databases

A persistent memory layer for AI agents that consolidates related facts over time, ages unused memories on a forgetting-curve model, and links memories through a shared entity graph instead of storing every fact as an undifferentiated vector.

• Agent memory

A desktop AI agent that runs tasks locally through its own sandboxed execution runtime, connecting to a model of your choice and using skills and MCP tools to read files, run commands and automate scheduled work from a chat interface.

• Agent frameworks

A self-hosted, read-only MCP server that lets an AI assistant answer questions about your own bank accounts by connecting through the Enable Banking open-banking provider, without storing balances or transactions itself.

• MCP servers
• APIs

A multi-agent AI infrastructure framework that adds scheduling, memory, human approval, cost control and security as reusable runtime modules around agent tool calls, rather than just letting a model call tools.

• Agent frameworks

A reinforcement learning framework built on Apple's MLX array framework, with single-file implementations of algorithms such as DQN, PPO and SAC that run end-to-end on device, including the environments, for over a million steps per second on Apple silicon.

• Models

A set of ROS 2 packages that bring llama.cpp and llava.cpp into robotics projects, letting a robot run GGUF language and vision-language models, including grammar-constrained output and real-time LoRA adapter changes.

• Inference

A personal, mostly Chinese-language PostgreSQL knowledge base by a PostgreSQL DBA, collecting articles and write-ups on internals, DBA practice, performance tuning, and production troubleshooting.

• Databases
• Curated lists

A free security-scanning skill for Claude Code, Codex, and Antigravity that traces data flow through AI-written code to find and explain 21 common vulnerability types.

• Code quality
• Security

A terminal AI coding agent focused on token efficiency, using cached multi-phase agents and a tree-sitter based virtual filesystem to read and edit minified code, plus programmable JSON pipelines and a visual plan-review mode.

• Coding agents

A native C++ MCP plugin that gives AI assistants full control of Unreal Engine 5.7/5.8 through a handful of namespace-dispatch tools instead of thousands of individual tool definitions.

• MCP servers