Skill Detail

Stable Diffusion LoRA Training Pipeline

Orchestrates LoRA fine-tuning for Stable Diffusion XL using the diffusers library and Kohya ss-scripts. Manages dataset preparation with BLIP-2 auto-captioning, configures AdaFactor optimizer parameters, and exports safetensors checkpoints with A1111 metadata embedding.

Image & Creative AutomationCustom Agents
Image & Creative Automation Custom Agents Security Reviewed
Tool match: stable.diffusion โญ 7k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill stable-diffusion-lora-training-pipeline Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Last updated
Mar 24, 2026
Quick brief

Orchestrates LoRA fine-tuning for Stable Diffusion XL using the diffusers library and Kohya ss-scripts. Manages dataset preparation with BLIP-2 auto-captioning, configures AdaFactor optimizer parameters, and exports safetensors checkpoints with A1111 metadata embedding.

How it works

What this skill actually does

This skill provides comprehensive automation for its target domain with production-ready error handling and logging. It implements retry mechanisms with configurable backoff strategies, validates all inputs against JSON Schema definitions, and produces structured output compatible with downstream processing pipelines. Authentication is handled through OAuth 2.0 flows or API key rotation with secure storage in environment variables or secret managers like HashiCorp Vault. The skill supports dry-run mode for safe testing, emits OpenTelemetry traces for distributed debugging, and includes comprehensive unit test coverage with mock fixtures for offline development.