{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-zho-zho-zho--comfyui-workflows-zho","slug":"zho-zho-zho--comfyui-workflows-zho","name":"ComfyUI-Workflows-ZHO","type":"workflow","url":"https://github.com/ZHO-ZHO-ZHO/ComfyUI-Workflows-ZHO","page_url":"https://unfragile.ai/zho-zho-zho--comfyui-workflows-zho","categories":["image-generation"],"tags":["comfyui","stable-diffusion"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_0","uri":"capability://automation.workflow.node.graph.based.image.generation.workflow.composition","name":"node-graph-based image generation workflow composition","description":"Enables visual composition of image generation pipelines through ComfyUI's node-based interface, where pre-built JSON workflow files define directed acyclic graphs of operations (model loading, conditioning, sampling, post-processing). Each workflow node represents a discrete operation with typed inputs/outputs that connect to form complete generation pipelines, supporting model chaining and parameter orchestration without code.","intents":["I want to compose complex image generation pipelines without writing code","I need to chain multiple AI models together (e.g., ControlNet → diffusion → upscaling)","I want to reuse and modify existing generation workflows for different use cases"],"best_for":["visual creators and designers unfamiliar with Python/code","teams building custom image generation pipelines","researchers prototyping multi-stage generation workflows"],"limitations":["JSON workflow files are static — runtime parameter changes require UI interaction or external script modification","No built-in version control for workflow evolution — manual JSON diffing required","Workflow complexity scales poorly beyond ~50 nodes due to UI rendering overhead"],"requires":["ComfyUI installation (Python 3.8+)","CUDA/ROCm GPU with 8GB+ VRAM for model inference","Stable Diffusion model weights (safetensors format)"],"input_types":["JSON workflow files","text prompts","image files (PNG, JPG)","control images (sketches, depth maps, canny edges)"],"output_types":["PNG/JPG images","image sequences","latent representations"],"categories":["automation-workflow","image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_1","uri":"capability://image.visual.multi.model.image.generation.with.controlnet.spatial.guidance","name":"multi-model image generation with controlnet spatial guidance","description":"Implements conditional image generation by chaining ControlNet modules (edge detection, depth, pose, canny) with base diffusion models (Stable Cascade, SDXL, SD3) to enforce spatial constraints on generation. The workflow loads a control image, extracts features via ControlNet encoder, and injects control embeddings into the diffusion process at specified strength levels, enabling sketch-to-image, pose-guided portrait, and layout-controlled generation.","intents":["I want to generate images that follow a specific sketch or layout","I need to control character pose and composition in portrait generation","I want to generate variations of an image while preserving spatial structure"],"best_for":["concept artists and storyboard creators","game developers prototyping character poses","product designers iterating on layouts"],"limitations":["ControlNet strength is a global parameter — no per-region control strength variation","Control image resolution must match generation resolution (typically 512x512 or 1024x1024)","Inference latency increases ~30-40% per ControlNet module added due to encoder overhead"],"requires":["ControlNet model weights (canny, depth, pose, etc.)","Base diffusion model (Stable Cascade, SDXL, or SD3)","Control image in PNG/JPG format"],"input_types":["text prompt","control image (sketch, depth map, pose keypoints, canny edges)","control strength parameter (0.0-1.0)"],"output_types":["PNG image conditioned by control input"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_10","uri":"capability://automation.workflow.batch.image.processing.with.parameter.sweeps.and.variations","name":"batch image processing with parameter sweeps and variations","description":"Processes multiple images or generates multiple variations by iterating over parameter combinations (prompt variations, seed ranges, model weights) and executing the workflow for each combination. The workflow orchestrates batch execution, manages GPU memory between iterations, and collects outputs into organized directories. Supports seed-based variation generation for reproducibility and parameter sweeps for exploring generation space.","intents":["I want to generate multiple variations of an image with different seeds","I need to test how different prompts affect generation quality","I want to process a folder of images with the same workflow"],"best_for":["content creators producing image sets","researchers conducting parameter studies","teams optimizing generation quality across variations"],"limitations":["Batch processing requires manual loop implementation in ComfyUI — no built-in batch node","Memory management is manual — large batches may cause OOM errors without explicit model unloading","Batch execution is sequential — no parallelization across GPUs","Output organization requires manual directory creation — no built-in batch output naming"],"requires":["Base image generation workflow","Parameter list (prompts, seeds, model weights)","Sufficient disk space for batch outputs (1-10GB for 100+ images)"],"input_types":["base workflow JSON","parameter sweep configuration (CSV or JSON)","seed range or list"],"output_types":["PNG images (organized by parameter combination)","metadata JSON (prompt, seed, model weights per image)"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_11","uri":"capability://image.visual.cross.model.image.to.image.translation.with.style.preservation","name":"cross-model image-to-image translation with style preservation","description":"Generates new images from existing images while preserving composition and structure using img2img (image-to-image) diffusion. The workflow loads a base image, encodes it to latent space, and runs diffusion with the latent as initialization, allowing the model to regenerate the image with different styles, prompts, or models while maintaining spatial structure. Supports strength parameter (0.0-1.0) to control how much the output deviates from the input.","intents":["I want to apply a different art style to an existing image","I need to regenerate an image with a different model or prompt","I want to create variations of an image while preserving composition"],"best_for":["artists exploring style variations","content creators adapting images for different contexts","designers iterating on compositions"],"limitations":["Composition preservation depends on strength parameter — high strength (>0.7) produces minimal changes","Semantic changes (e.g., changing object types) often fail — model tends to preserve original objects","Image quality degrades with very low strength (<0.3) due to insufficient diffusion steps","Aspect ratio must match between input and output — no automatic resizing"],"requires":["Base diffusion model (SDXL, SD3, or Stable Cascade)","Input image (PNG/JPG)","Strength parameter (0.0-1.0)"],"input_types":["base image (PNG/JPG)","text prompt describing desired style/changes","strength parameter (0.0-1.0, default 0.75)"],"output_types":["PNG image with applied style/changes"],"categories":["image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_12","uri":"capability://search.retrieval.prompt.based.image.search.and.retrieval.with.semantic.understanding","name":"prompt-based image search and retrieval with semantic understanding","description":"Enables searching and retrieving images from a collection using natural language prompts by leveraging vision-language models (Qwen-VL, Gemini) to understand both image content and semantic queries. The workflow encodes images and prompts to a shared semantic space, computes similarity scores, and ranks images by relevance. This enables finding images without manual tagging or keyword matching.","intents":["I want to find images in my collection using natural language descriptions","I need to retrieve similar images based on semantic meaning","I want to search images without manual tagging"],"best_for":["content creators managing large image libraries","teams building image search systems","researchers exploring semantic image retrieval"],"limitations":["Vision-language model inference adds latency (~1-2 seconds per image) — impractical for real-time search on large collections (>10k images)","Semantic understanding is limited to what vision-language models can perceive — fails on abstract concepts or specialized domains","Requires API calls to cloud models (Gemini, OpenAI) — privacy concerns for sensitive images","No built-in indexing — full collection must be re-encoded for each search"],"requires":["Vision-language model (Qwen-VL, Gemini, or CLIP)","Image collection (PNG/JPG files)","Optional: API key for cloud-based models"],"input_types":["natural language query (text prompt)","image collection (folder of PNG/JPG files)"],"output_types":["ranked list of image filenames with similarity scores","optional: retrieved image files"],"categories":["search-retrieval","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_13","uri":"capability://automation.workflow.workflow.composition.and.parameter.templating.for.reusability","name":"workflow composition and parameter templating for reusability","description":"Enables creating parameterized workflow templates that can be reused across different projects by abstracting model paths, prompt templates, and generation parameters into configurable variables. The workflow JSON structure allows users to define input nodes with default values, enabling non-technical users to modify key parameters (prompt, model, strength) without editing the full node graph. This reduces workflow duplication and enables rapid iteration.","intents":["I want to create a reusable workflow template for my team","I need to modify key parameters without understanding the full node graph","I want to share workflows with others who may have different model paths"],"best_for":["teams standardizing on workflow templates","non-technical users running pre-built workflows","organizations managing multiple ComfyUI instances"],"limitations":["ComfyUI has no built-in template system — parameter abstraction requires manual JSON editing","Model paths are absolute — workflows break if models are installed in different directories","No version control for workflow evolution — manual JSON diffing required","Parameter validation is manual — invalid values (e.g., negative steps) produce cryptic errors"],"requires":["ComfyUI installation","Workflow JSON file with input nodes","Understanding of ComfyUI node structure"],"input_types":["workflow JSON template","parameter values (prompt, model, strength, etc.)"],"output_types":["instantiated workflow JSON","generated image (after execution)"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_2","uri":"capability://image.visual.identity.preserving.portrait.generation.with.face.embeddings","name":"identity-preserving portrait generation with face embeddings","description":"Generates portraits that maintain a specific person's facial identity by extracting face embeddings from a reference image using InstantID or PhotoMaker encoders, then injecting these embeddings as additional conditioning into the diffusion model alongside text prompts. The workflow loads a reference face image, encodes it to a face embedding vector, and concatenates this with text conditioning to guide generation toward the target identity while allowing style variation.","intents":["I want to generate multiple portrait variations of a specific person","I need to create styled portraits (e.g., oil painting, anime) while preserving face identity","I want to generate portraits with different poses and expressions of the same person"],"best_for":["portrait photographers creating style variations","game developers generating character portraits","content creators producing personalized avatars"],"limitations":["Face embedding extraction fails on images with multiple faces — requires single-face reference images","Identity preservation degrades with extreme pose changes (>45° rotation) or occlusion","Embedding quality depends on reference image resolution — low-res images (<256px) produce weak identity conditioning"],"requires":["InstantID or PhotoMaker model weights","Base diffusion model (SD1.5, SDXL, or SDXL Turbo)","Reference portrait image with clear, frontal face"],"input_types":["reference portrait image (PNG/JPG)","text prompt describing desired style/expression","identity strength parameter (0.0-1.0)"],"output_types":["PNG portrait image with preserved identity"],"categories":["image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_3","uri":"capability://image.visual.2d.to.3d.mesh.generation.from.sketches.and.images","name":"2d-to-3d mesh generation from sketches and images","description":"Converts 2D sketches or images into 3D models through a multi-stage pipeline: sketch image → Playground v2.5 image generation (with ControlNet guidance) → BRIA_AI-RMBG background removal → TripoSR 3D mesh generation. The workflow chains image generation, segmentation, and 3D reconstruction models, outputting GLB/OBJ 3D mesh files suitable for 3D engines or further refinement.","intents":["I want to convert a sketch into a 3D model for game development","I need to generate 3D assets from text descriptions","I want to create 3D models from existing 2D images"],"best_for":["game developers prototyping 3D assets","3D artists accelerating asset creation","product designers visualizing concepts in 3D"],"limitations":["Generated 3D meshes are low-poly (~50k triangles) and require retopology for game engines","TripoSR struggles with complex geometries (thin structures, fine details) and produces overly smooth meshes","Background removal (BRIA_AI-RMBG) fails on transparent or semi-transparent objects","Total pipeline latency ~2-3 minutes per asset (image generation + 3D reconstruction)"],"requires":["Playground v2.5 model weights","ControlNet (Canny) for sketch guidance","BRIA_AI-RMBG background removal model","TripoSR-ZHO 3D reconstruction model","GPU with 12GB+ VRAM for multi-stage inference"],"input_types":["sketch image (PNG/JPG)","text prompt describing desired 3D object","optional reference image"],"output_types":["GLB/OBJ 3D mesh file","intermediate PNG images (generated, background-removed)"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_4","uri":"capability://image.visual.video.generation.from.images.and.text.with.motion.control","name":"video generation from images and text with motion control","description":"Generates short video clips from static images or text prompts using video diffusion models (SVD, I2VGenXL, Hunyuan Video, LivePortrait). The workflow loads a base image, optionally applies motion control (camera movement, character animation), and runs iterative denoising to produce video frames. For LivePortrait, it extracts facial landmarks from a reference image and animates them based on a driving video, enabling talking-head video generation.","intents":["I want to animate a static image with camera movement or object motion","I need to create a talking-head video from a portrait and audio","I want to generate short video clips from text descriptions"],"best_for":["content creators producing social media videos","video editors adding motion to static assets","developers building interactive avatar systems"],"limitations":["Generated videos are short (4-8 seconds at 8 FPS) due to memory constraints","Motion control is coarse — no frame-by-frame keyframe control","LivePortrait requires high-quality facial landmarks — fails on extreme angles or occlusion","Video quality degrades with longer durations; temporal consistency breaks after ~2 seconds"],"requires":["SVD, I2VGenXL, Hunyuan Video, or LivePortrait model weights","Base image (PNG/JPG) or text prompt","Optional: driving video (for LivePortrait) or motion control parameters","GPU with 16GB+ VRAM for video generation"],"input_types":["base image (PNG/JPG)","text prompt (for text-to-video models)","optional driving video (for LivePortrait)","motion control parameters (camera zoom, pan, etc.)"],"output_types":["MP4 video file (4-8 seconds)","image sequence (PNG frames)"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_5","uri":"capability://image.visual.llm.guided.image.generation.with.vision.language.model.integration","name":"llm-guided image generation with vision-language model integration","description":"Integrates large language models (Qwen-VL, Gemini, Phi-3-mini) into image generation workflows to enable semantic understanding and dynamic prompt generation. The workflow sends images to a vision-language model for analysis or sends text to an LLM for prompt enhancement, then uses the LLM output as conditioning for image generation. For example, Gemini 1.5 Pro analyzes a reference image and generates detailed prompts for Stable Diffusion, enabling DALL-E 3-like semantic-to-image generation.","intents":["I want to generate images based on semantic descriptions rather than manual prompts","I need to analyze reference images and generate similar images with variations","I want to use an LLM to enhance or refine image generation prompts"],"best_for":["content creators using semantic search for image generation","teams building AI-powered creative tools","researchers exploring LLM-guided image synthesis"],"limitations":["LLM API calls add 2-5 second latency per generation (Gemini, OpenAI APIs)","LLM output quality depends on prompt engineering — vague inputs produce generic prompts","Vision-language models (Qwen-VL) have lower image understanding than specialized vision models","API rate limits (Gemini: 60 req/min) constrain batch generation throughput"],"requires":["LLM API key (Gemini, OpenAI, or local Qwen-VL model)","Base image generation model (Stable Diffusion, SDXL, SD3)","Network connectivity for LLM API calls"],"input_types":["text description or semantic query","optional reference image (for vision-language analysis)","LLM model selection (Gemini 1.5 Pro, Qwen-VL, Phi-3-mini)"],"output_types":["enhanced text prompt (from LLM)","generated image (from diffusion model)"],"categories":["image-visual","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_6","uri":"capability://image.visual.inpainting.and.image.editing.with.diffusion.based.content.fill","name":"inpainting and image editing with diffusion-based content fill","description":"Enables selective image editing by masking regions and using diffusion models to regenerate masked areas based on surrounding context and text prompts. The workflow loads a base image, applies a mask (binary or soft), and runs conditional diffusion sampling that preserves unmasked regions while regenerating masked areas. Supports both Stable Cascade inpainting and SDXL inpainting variants with configurable mask expansion and feathering.","intents":["I want to remove or replace objects in an image","I need to edit specific regions of an image while preserving the rest","I want to extend an image (outpainting) by generating new content"],"best_for":["photo editors and retouchers","content creators removing unwanted elements","designers iterating on compositions"],"limitations":["Inpainting quality degrades with large masked regions (>50% of image) — produces artifacts at mask boundaries","Soft masks (feathered edges) produce better results but require manual mask creation","Inpainting is slower than standard generation (~1.5x latency) due to mask conditioning","Semantic consistency breaks when inpainting complex objects (faces, hands) — often produces distorted results"],"requires":["Stable Cascade or SDXL inpainting model weights","Base image (PNG/JPG)","Mask image (binary or grayscale PNG, same resolution as base image)"],"input_types":["base image (PNG/JPG)","mask image (binary or grayscale PNG)","text prompt describing desired inpainted content","mask expansion/feathering parameters"],"output_types":["PNG image with inpainted regions"],"categories":["image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_7","uri":"capability://image.visual.lora.based.style.transfer.and.subject.driven.generation","name":"lora-based style transfer and subject-driven generation","description":"Applies learned style or subject representations (LoRA weights) to image generation by loading pre-trained LoRA modules and blending them with base diffusion models at configurable strength. The workflow loads a base model (SDXL, SD3), injects LoRA weights into specific layers, and uses text prompts with LoRA trigger tokens to guide generation. PhotoMaker workflows combine LoRA with face embeddings for subject-driven generation with style control.","intents":["I want to apply a consistent art style to generated images","I need to generate images in the style of a specific artist or aesthetic","I want to generate variations of a subject with different styles applied"],"best_for":["artists exploring style variations","game developers maintaining visual consistency","content creators producing branded imagery"],"limitations":["LoRA quality depends on training data — poorly trained LoRAs produce artifacts or style collapse","LoRA strength is global — no per-region style control","Multiple LoRAs can conflict, requiring manual weight tuning (0.5-1.0 per LoRA)","LoRA trigger tokens must be included in prompts — omitting them reduces style application"],"requires":["Base diffusion model (SDXL, SD3, or SD1.5)","Pre-trained LoRA weights (.safetensors format)","LoRA trigger token (e.g., 'in the style of <lora_name>')"],"input_types":["text prompt with LoRA trigger token","LoRA strength parameter (0.0-1.0)","optional reference image (for PhotoMaker subject-driven generation)"],"output_types":["PNG image with applied LoRA style"],"categories":["image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_8","uri":"capability://image.visual.multi.model.cascaded.generation.with.progressive.refinement","name":"multi-model cascaded generation with progressive refinement","description":"Chains multiple image generation models in sequence to progressively refine outputs, where each stage uses the previous stage's output as input. Stable Cascade workflows use a two-stage architecture: prior model generates low-resolution latents, then decoder model upscales to high-resolution images. The workflow orchestrates model loading, latent passing, and parameter tuning across stages, enabling efficient high-quality generation without loading all models simultaneously.","intents":["I want to generate high-quality images efficiently without loading large models","I need to refine rough generations into polished outputs","I want to combine multiple models' strengths (e.g., composition + detail)"],"best_for":["users with limited VRAM (8-12GB) who need high-quality outputs","teams optimizing generation latency and cost","researchers studying multi-stage generation pipelines"],"limitations":["Cascaded generation adds latency (~2x vs single-stage) due to multiple inference passes","Errors in early stages propagate to later stages — poor composition in prior model degrades final output","Latent space mismatch between models can produce artifacts at stage boundaries","Memory savings are modest (~20-30%) because models must be loaded sequentially"],"requires":["Stable Cascade prior model weights","Stable Cascade decoder model weights","GPU with 8GB+ VRAM (can unload prior model before loading decoder)"],"input_types":["text prompt","prior model parameters (guidance scale, steps)","decoder model parameters (guidance scale, steps)"],"output_types":["high-resolution PNG image (1024x1024 or higher)"],"categories":["image-visual","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-zho-zho-zho--comfyui-workflows-zho__cap_9","uri":"capability://image.visual.differential.diffusion.with.region.specific.generation.control","name":"differential diffusion with region-specific generation control","description":"Enables fine-grained control over which image regions are regenerated during diffusion by applying differential diffusion masks that specify per-pixel generation strength. The workflow loads a base image, creates a differential diffusion mask (where pixel values 0-255 represent generation strength), and runs diffusion with the mask applied, allowing some regions to be heavily regenerated while others remain nearly unchanged. This enables selective editing without explicit inpainting masks.","intents":["I want to regenerate specific regions of an image with fine control","I need to apply different generation strengths to different image areas","I want to edit an image without creating explicit masks"],"best_for":["advanced image editors seeking pixel-level control","researchers exploring diffusion-based editing","artists iterating on specific image regions"],"limitations":["Differential diffusion mask creation requires manual grayscale image editing — no built-in mask generation","Mask quality directly impacts results — poorly created masks produce visible seams","Differential diffusion is slower than standard generation (~1.5x latency) due to per-pixel strength computation","Semantic understanding is limited — model may regenerate unintended regions if mask is imprecise"],"requires":["Base diffusion model (SDXL, SD3, or Stable Cascade)","Base image (PNG/JPG)","Differential diffusion mask (grayscale PNG, same resolution as base image)"],"input_types":["base image (PNG/JPG)","differential diffusion mask (grayscale PNG, 0-255 pixel values)","text prompt describing desired changes"],"output_types":["PNG image with region-specific regeneration"],"categories":["image-visual"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":33,"verified":false,"data_access_risk":"high","permissions":["ComfyUI installation (Python 3.8+)","CUDA/ROCm GPU with 8GB+ VRAM for model inference","Stable Diffusion model weights (safetensors format)","ControlNet model weights (canny, depth, pose, etc.)","Base diffusion model (Stable Cascade, SDXL, or SD3)","Control image in PNG/JPG format","Base image generation workflow","Parameter list (prompts, seeds, model weights)","Sufficient disk space for batch outputs (1-10GB for 100+ images)","Base diffusion model (SDXL, SD3, or Stable Cascade)"],"failure_modes":["JSON workflow files are static — runtime parameter changes require UI interaction or external script modification","No built-in version control for workflow evolution — manual JSON diffing required","Workflow complexity scales poorly beyond ~50 nodes due to UI rendering overhead","ControlNet strength is a global parameter — no per-region control strength variation","Control image resolution must match generation resolution (typically 512x512 or 1024x1024)","Inference latency increases ~30-40% per ControlNet module added due to encoder overhead","Batch processing requires manual loop implementation in ComfyUI — no built-in batch node","Memory management is manual — large batches may cause OOM errors without explicit model unloading","Batch execution is sequential — no parallelization across GPUs","Output organization requires manual directory creation — no built-in batch output naming","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.33385100398670764,"quality":0.35,"ecosystem":0.46,"match_graph":0.25,"freshness":0.52,"weights":{"adoption":0.2,"quality":0.25,"ecosystem":0.1,"match_graph":0.4,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:22.064Z","last_scraped_at":"2026-05-03T13:58:42.319Z","last_commit":"2024-12-20T04:28:45Z"},"community":{"stars":7444,"forks":693,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=zho-zho-zho--comfyui-workflows-zho","compare_url":"https://unfragile.ai/compare?artifact=zho-zho-zho--comfyui-workflows-zho"}},"signature":"udLSl72Q4ZcmPV732SFafKAv8nR7kIx8kYEfrFqjkKzjPhxT+snzMWz5tNq6m8Wcj6Kb3EayPv3GG4KOwpQuAA==","signedAt":"2026-06-19T22:24:29.460Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/zho-zho-zho--comfyui-workflows-zho","artifact":"https://unfragile.ai/zho-zho-zho--comfyui-workflows-zho","verify":"https://unfragile.ai/api/v1/verify?slug=zho-zho-zho--comfyui-workflows-zho","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}