Skill Creator
atrawog
41 published skills · 0 GitHub stars across the catalog. View on GitHub ↗
gpu
GPU driver configuration for Bazzite. NVIDIA proprietary drivers, Optimus laptops,NVK (open-source NVIDIA), GPU switching, Broadcom WiFi, and Mesa testing builds.Use when users need to configure graphics drivers.
network
Network configuration for Bazzite. iwd WiFi backend, Wake-on-LAN, and Tailscale VPN.Use when users need to configure network services. For SSH, see bazzite-ai:configure.
security
Security configuration for Bazzite. LUKS disk encryption with TPM auto-unlock,secure boot key management, and sudo password feedback. Use when users needto configure security features.
storage
Storage management for Bazzite. Automounting drives (BTRFS/EXT4, Framework, SteamOS),BTRFS deduplication, rmlint disk trimming, and snapper snapshots. Use when usersneed to configure disk and partition management.
system
System maintenance for Bazzite OS. Updates via topgrade, cleanup of podman/flatpaks,viewing logs and changelogs, diagnostics, and power measurements. Use when usersneed to update, clean, or diagnose their Bazzite system.
virtualization
GPU passthrough and virtualization for Bazzite. KVM/VFIO setup, Looking Glass (kvmfr),USB hotplug for VMs, and libvirt configuration. Use when users need GPU passthroughor advanced virtualization features.
reward
Reward model training for RLHF pipelines. Covers RewardTrainer, preference datasetpreparation, sequence classification heads, and reward scaling for stablereinforcement learning. Includes thinking quality scoring patterns.
rloo
Reinforcement Learning with Leave-One-Out estimation for policy optimization.Covers RLOOTrainer, reward function integration, baseline estimation, andvariance reduction techniques for stable RL training. Includes thinking-aware patterns.
openwebui
Open WebUI AI chat interface management via Podman Quadlet. Provides a web UIfor interacting with Ollama models. Use when users need to configure, start,or manage the Open WebUI service.
dpo
Direct Preference Optimization for learning from preference pairs. Covers DPOTrainer,preference dataset preparation, implicit reward modeling, and beta tuning forstable preference learning without explicit reward models. Includes thinking quality patterns.
qlora
Advanced QLoRA experiments and comparisons. Covers alpha scaling, LoRA rank selection,target module strategies, continual learning, multi-adapter hot-swapping, andquantization comparison (4-bit vs BF16).
vision
Vision model fine-tuning with FastVisionModel. Covers Pixtral, Ministral VL training,UnslothVisionDataCollator, image+text datasets, and vision-specific LoRA configuration.
grpo
Group Relative Policy Optimization for reinforcement learning from human feedback.Covers GRPOTrainer, reward function design, policy optimization, and KL divergenceconstraints for stable RLHF training. Includes thinking-aware reward patterns.
inference
Fast inference with Unsloth and vLLM backend. Covers model loading, fast_generate(),thinking model output parsing, and memory management for efficient inference.
sft
Supervised Fine-Tuning with SFTTrainer and Unsloth. Covers dataset preparation,chat template formatting, training configuration, and Unsloth optimizationsfor 2x faster instruction tuning. Includes thinking model patterns.
localai
LocalAI local inference API management via Podman Quadlet. Provides anOpenAI-compatible API for local model inference with GPU acceleration.Use when users need to configure, start, or manage the LocalAI service.
config
Unified system configuration dispatcher for bazzite-ai. Manages services(Docker, Cockpit, SSH), desktop settings (gamemode, Steam), security(passwordless sudo), and development environment (GPU containers). Usewhen users need to enable/disable system features or check configuration status.