agentsclimarketplace

Comfyui workflow

Skill yigityildiz0/universal-ai-skill-library/skills/common/comfyui-workflow

531 searchable AI Agent Skills for Claude Code, OpenAI Codex, and OpenCode — EN/TR catalog, platform and risk notes, direct ZIPs, and curated bundles.

Install
npx -y skills add yigityildiz0/universal-ai-skill-library --skill comfyui-workflow

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

  • 20 days oldThe repository was created 20 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Generate and debug ComfyUI workflow JSON files for FLUX, SDXL, SD3, Wan2.2, img2img, txt2img, txt2vid, img2vid, LoRA, ControlNet, IPAdapter, inpaint, and upscale tasks. Use when the user wants to build, modify, compare, or troubleshoot ComfyUI workflows.

SKILL.md

2.9 KB, 667 tokens by cl100k_base, as published. Nobody here has run it

ComfyUI Workflow

Use this skill when the user wants to create, modify, compare, or troubleshoot a ComfyUI workflow.

Goals

  • Generate valid ComfyUI workflow JSON files
  • Start from the closest template in templates/ whenever possible
  • Use references/ to check node names, inputs, outputs, defaults, and constraints
  • Keep explanations short and practical
  • Prefer working solutions over theory

What to gather first

Before generating a workflow, identify:

  1. Task type

    • txt2img
    • img2img
    • txt2vid
    • img2vid
    • inpaint
    • upscale
    • control workflow
    • identity-preserving edit

  2. Model family

    • FLUX
    • SDXL
    • SD3
    • Wan 2.2
    • HunyuanVideo
    • LTXV
    • Mochi
    • Cosmos
    • other

  3. Key settings

    • resolution
    • steps
    • CFG
    • sampler
    • seed behavior

  4. Optional modules

    • LoRA
    • ControlNet
    • IPAdapter
    • PuLID
    • upscaler
    • face/detail nodes

  5. Prompt inputs

    • positive prompt
    • negative prompt
    • trigger words if a LoRA is used

If the user is vague, suggest a sane default configuration.

Workflow generation rules

  • Use the closest file in templates/ as the base whenever possible
  • Adapt the template instead of building everything from zero unless necessary
  • Output valid ComfyUI LiteGraph UI JSON
  • Include a short explanation of the pipeline
  • List required custom nodes
  • List required model files
  • Mention likely missing dependencies if relevant

Debugging rules

When the user shares a broken workflow, console error, or screenshot, troubleshoot in this order:

  1. Missing model files
  2. Missing custom nodes
  3. Wrong node connections
  4. Wrong widget values
  5. Incompatible model family
  6. VAE / text encoder mismatch
  7. LoRA placement or wrong trigger usage
  8. ControlNet / IPAdapter / PuLID integration mistakes
  9. Unsupported or outdated node names
  10. Resolution / VRAM / precision issues

FLUX-specific checks

For FLUX workflows, always verify:

  • correct model loader path
  • text encoder setup
  • VAE compatibility
  • LoRA attachment points
  • filename consistency
  • sampler path
  • output node presence

Output style

  • Be concise
  • Give the final workflow first if the request is clear
  • Then list required models and custom nodes
  • If debugging, first state the most likely cause
  • Then give the fastest fix
  • Then give the clean long-term fix

Examples

  • Generate a FLUX txt2img workflow with LoRA support
  • Build a Wan 2.2 img2vid workflow with camera control
  • Debug this ComfyUI workflow JSON
  • Add PuLID and identity-preserving editing to this FLUX workflow
  • Compare two workflows and explain which one is more stable

What ships with it: 80 files

663.3 KB alongside SKILL.md

comfyui-workflow/

40 more files not listed here. See all 80 in the repository.

Keep looking

Skills are one crate of 327,132. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.