Art & Design

Codex local skill for distinctive production-grade frontend design

Brand-pluggable design system for Coding Assistans (Claude Code, Gemini CLI, CoPilot CLI) - turn prompts or markdown into brand-perfect PowerPoint decks. Marketplace of one plugin (more coming).

Public Codex skills for browser-game Sentinel playtesting, UI review, game-design critique, safe learning, and skill packaging.

The most intelligent context optimization engine for coding agents. Code-aware AST parsing across 17 languages. Command rewriting. Test, build, and git output compression. Token budget cascading. Built in Rust.

Learn Nanobot with a beginner-friendly guide to AI Agent basics, interview prep, and core concepts for job seekers

AI Agent Development Hub - Comprehensive collection of agent frameworks, tools, and production-ready projects for building intelligent agents. From foundational frameworks like LangChain to complete agent applications.

AgentBoard is opensource TUI based Kanban board for orchestrating AI coding agents

Verifiable audit substrate for AI agents — EU AI Act Article 12 ready. Apache 2.0.

ATMOS - Atmosphere for Agentic Builders

Build a local coding agent with Ollama, workspace snapshots, tool use, approvals, and persistent memory for safer automated coding

A modular, extensible AI agent built on top of adk-rust and the teloxide framework.

Experimental coding agent with a Textual TUI and a single run_python action surface.

Generate project-specific design system rules for Figma-to-code workflows

A lightweight terminal coding assistant with Claude Code-like workflow, tool loop, and TUI architecture, built for learning and experimentation. Multi-language support: TypeScript , Python and Rust implementations available now.

Convert flat logos into spinning 3D coins using a single command.

Self-hosted AI agent with configurable system access, designed to run on a budget VPS (~$10/month for AI + infrastructure). Built for a single user as a personal automation platform controlled via Telegram.

Local Pokemon Red LLM-agent harness with live gameplay UI, structured traces, save states, replay, and turn-by-turn agent observability.

Easiest and laziest way for building multi-agent LLMs applications.

Autonomous productivity agents built with Apache Airflow + NVIDIA NemoClaw. Two pipelines: NYC 311 overnight triage (saves 47 min/day for city agency supervisors) and model migration eval (replaces 2-3 days of manual LLM comparison). Airflow is the schedule. NemoClaw is the brain.

Arcana Skills is a practical ecosystem of installable domain packages for real-world AI work. Each package is built as a routed system of linked mini-skills, references, examples, and support files, designed for selective loading, structural integrity, and actual execution rather than theory alone.

An AI-native software engineering methodology — phased LLM agents for architecture, domain design, and planning before dev starts.

Colony: build and deploy autonomous LLM agents

GraphBit is the world’s first enterprise-grade Agentic AI framework, built on a Rust core with a Python wrapper for unmatched speed, security, and scalability. It enables reliable multi-agent workflows with minimal CPU and memory usage, making it production-ready for real-world enterprise environments.

TypeScript SDK for server-run agent loops with resumable client-side tools, MCP integration, AG-UI adapters, and durable protocol contracts.