Hands-on examples for the NVIDIA NeMo Agent Toolkit (NAT) with Ollama as local inference backend - ReAct agents, custom Python tools, and multi-agent orchestration patterns on a single GPU workstation.
npx add-skill custom-build-robots/nemo-agent-toolkit-examplesHands-on examples for the NVIDIA NeMo Agent Toolkit (NAT) with Ollama as local inference backend - ReAct agents, custom Python tools, and multi-agent orchestration patterns on a single GPU workstation.
Source: https://github.com/custom-build-robots/nemo-agent-toolkit-examples
Discovered from GitHub repositories pushed in the last 24 hours for agent skills, Claude/Codex/Gemini workflows, MCP tooling, and adjacent AI-agent automation.
OpenCode Design System Toolkit: design-md-wizard agent, /design-md-init command, and design-md-apply skill — a unified set of extensions that creates, validates, and enforces DESIGN.md from surveys, codebases, or Figma tokens, and automatically syncs the design system reference into AGENTS.md and CLAUDE.md for consistent, on-brand UI generation.
Codex skill for image generation via configurable compatible image APIs
Claude Code skill: Export Mac WeChat chat to Obsidian with image decryption
Claude Code skill + /codex-cli-execution slash command — Claude supervises Codex CLI in interactive --yolo mode via tmux: shapes prompt, pastes, polls TUI for idle, surfaces errors, evaluates result, delegates corrections.
Extension skills for the superpowers suite: regression testing, pre-push review, refactor analysis, decision tracking, Roslyn code graph integration, and companion ecosystem plugins for Claude Code
Claude Code native — feature guide skill, deterministic verify hook, LLM auditor agent. Keeps Claude Code feature knowledge fresh.