ComfyUI Auto Tagger for Eagle

ComfyUI Auto Tagger automatically extracts metadata from AI-generated images and organizes them into Eagle Tags and Notes.

Starting from version 1.3.2, the plugin provides comprehensive support not only for ComfyUI but also for Stable Diffusion WebUI (A1111/Forge) and Civitai generated images. It features advanced logic to parse complex node graphs, ensuring that even the most intricate workflows are accurately translated into a searchable library.

[Features]

• Multi-Format & Engine Support: Supports ComfyUI, A1111, and Civitai metadata across both PNG and WebP formats.

• Advanced Workflow Trace Logic: Unlike simple metadata readers, this plugin dynamically analyzes the ComfyUI node graph. It traces the actual execution path to identify parameters (Checkpoint, LoRA, Prompts, Seed, etc.) even in massive "All-in-One" workflows.

• Flexible Smart Tagging: Automatically converts prompts and parameters into searchable tags (e.g., #checkpoint_name, seed:12345) or detailed annotations in the Notes section.

• Batch Management: Process multiple images with a visual progress bar. Includes a Force Delete Mode (Shift + Click) to quickly clean up tags and notes added by the plugin.

• No External Dependencies: Operates by directly parsing image chunks, ensuring a lightweight and fast experience without heavy external libraries.

(Automatically converts prompts and parameters into Tags and Notes)

[Advanced Workflow Trace Logic (v1.3.2+)]

Unlike simple metadata readers, this plugin doesn't just "read" data; it dynamically analyzes the ComfyUI node graph to trace the actual execution path leading to the final image.

• Path-Based Extraction: It traces the latent and image chains back from the output nodes. This ensures that only the parameters from nodes that actually contributed to the image are extracted, filtering out inactive nodes or unused branches in complex "All-in-One" workflows.

• Multi-Stage Support: Comprehensive extraction of seeds, prompts, and samplers from all stages of generation, including KSampler, SAM Detailer, HiresFix, and FaceDetailer.

• Accurate Parameter Retrieval: Even with complex custom nodes like "Lora Loader Stack" or Reroute/Primitive nodes, the logic recursively resolves values to ensure accurate retrieval.

• Full Audit in Notes: Every detected generation step is recorded in detail within the Eagle Notes section, providing a precise "recipe" for reproduction.

(Smart Path Tracing: This logic intelligently navigates complex node graphs to identify and extract only the parameters that actually contributed to the final image.)

(Full Multi-Stage Audit: Simultaneously captures accurate generation data from every step, including KSampler, Detailers, and HiresFix, for a complete reproduction recipe.)

[Customizable Settings]

You have full control over your library’s organization. Toggle specific metadata items on or off to ensure only the information you need is imported:

• Output Destination: Independently enable/disable Tags and Notes.

• Extraction Targets: Selectively import Checkpoints, LoRAs, Prompts, Seeds, Steps, CFG, and Samplers.

• Debug Mode: Access detailed logs for troubleshooting complex custom nodes.

(Flexible settings to keep your library organized)

[Why use this?]

Eagle is excellent for managing assets, but AI generation creates a unique challenge: hidden metadata. 

This plugin turns that invisible data into a powerful, searchable database. With its unique Path-Based Extraction, it filters out inactive nodes and unused branches, giving you the most accurate "recipe" for every image in your collection.

[How to Use]

1. Select one or more AI-generated images in Eagle.

2. Right-click and choose Plugins > ComfyUI Auto Tagger.

3. Configure your output: choose whether to add to Tags, Notes, or both.

4. Select the specific metadata items you wish to extract (e.g., Checkpoint, LoRA, Sampler).

5. Click Start Tagging.