runcomfy-com/skills
πͺMarketplaceAgent skills for RunComfy β model-pinned brand skills and intent-routed category routers for the RunComfy Model API, distributed via skills.sh.
Add this Marketplace
/plugin marketplace add runcomfy-com/skillsThen open /plugin in Claude Code to pick and install its plugins.
Plugins in this Marketplace1
README.md
View on GitHub βRunComfy skills and plugins
Run ComfyUI workflows in the cloud on GPU, generate and edit images, create videos, and manage LoRA training. Use RunComfy as an AI image generator and AI video generator from your assistant, with AI image models and AI video models including FLUX, Seedream, Seedance, Wan, LTX, Nano Banana and Kling.
Cursor and Claude plugins
The RunComfy plugin bundles the hosted MCP endpoint with four focused skills for image generation, video generation, ComfyUI workflows and LoRA training. It supports Cursor and the shared Claude Code/Cowork plugin format. Marketplace approval is separate from publishing this source.
- Claude marketplace manifest:
.claude-plugin/marketplace.json - Cursor marketplace manifest:
.cursor-plugin/marketplace.json - Self-contained plugin:
plugins/runcomfy
See the plugin README for installation, authorization, example prompts and cost handling.
Individual CLI skills
The existing top-level skill directories provide task-specific and model-specific RunComfy CLI guidance. Install selected skills from this repository:
npx skills add runcomfy-com/skills --skill ai-image-generation
npx skills add runcomfy-com/skills --skill ai-video-generation
Browse existing skills on skills.sh. Some older skill documents retain historical installation links and model defaults; use this repository name and verify current model inputs and prices before running jobs.
RunComfy
AI image models and AI video models Β· ComfyUI workflows Β· AI Toolkit LoRA training Β· Documentation Β· Support
More from this repository10
Widens or extends a photo on RunComfy, a cloud image service, by adding matching scenery past its edges. Use it to change a picture's shape or fill in what the camera did not capture.
A skill that puts a different face or character into a photo or video with AI models on RunComfy. It picks the right model for your goal, such as a single still image or a moving clip.
Lets Claude create images from text with Flux 2 Klein, a fast image model from Black Forest Labs, on RunComfy. Use the small version for drafts in under a second and the larger one for detailed final images.
A skill that makes short videos (4 to 15 seconds) with sound from a text prompt, using the Seedance 2.0 Pro model on RunComfy. You can add reference photos, clips, and audio to keep a person or scene consistent.
A skill that removes or replaces one part of a photo, such as a watermark, a stray object, or the background. You mark the area with a black-and-white guide image, or just describe it in words.
A skill that makes music from a text description, from short jingles to full songs with lyrics. It can also redo one section of a track or make a track longer.
Makes a person's mouth in a video or photo move in time with an audio track, using AI models on RunComfy. You can dub an existing video, make a portrait talk or create a speaking video from a script.
Teaches your AI agent to use the RunComfy command-line tool to run hundreds of AI models. With one install and one sign-in, it can make images and videos, sync lips, swap faces and more.
Relight a still image on RunComfy, change lighting direction, color temperature, intensity, or mood without redoing the shot. Routes to Qwen Edit 2509's dedicated relight LoRA endpoint, with fallback to identity-preserving edit models for prose-based lighting descriptions.
A Claude Code skill for generating text-to-video with Wan-AI Wan 2.7 on RunComfy, featuring multi-reference conditioning, audio-driven lip-sync via audio_url, smoother transitions, and prompt expansion for high-quality video output.