Skill Detail

Stable Diffusion ControlNet Pose Pipeline

Orchestrates Stable Diffusion image generation with ControlNet pose conditioning via the Automatic1111 API. Chains OpenPose detection, depth estimation, and img2img endpoints.

Image & Creative AutomationCustom Agents
Image & Creative Automation Custom Agents Security Reviewed
Tool match: stable.diffusion โญ 33.8k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill stable-diffusion-controlnet-pose-pipeline Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Last updated
Apr 2, 2026
Quick brief

The Stable Diffusion ControlNet Pose Pipeline automates pose-conditioned image generation by orchestrating the Automatic1111 Stable Diffusion Web UI API. It chains multiple API endpoints starting with the /sdapi/v1/detect endpoint for OpenPose skeleton extraction from reference images, then feeds the detected pose maps into the /sdapi/v1/txt2img or /sdapi/v1/img2img endpoints with ControlNet extension parameters. The skill manages ControlNet model selection across pose, depth, canny, and normal map preprocessors, configuring control_mode, resize_mode, and guidance_start/end parameters for each conditioning layer. It supports multi-ControlNet setups where pose and depth maps are combined for more precise spatial control. Batch processing handles multiple reference images with consistent seed management for reproducible outputs. The skill integrates with the Civitai API to discover and download ControlNet models, and supports LoRA weight injection via the prompt syntax for style transfer. Output images are post-processed with the /sdapi/v1/extra-single-image upscaling endpoint using Real-ESRGAN or SwinIR models.