nodes/slahiri-ComfyUI-AI-Photography-Toolkit
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
102 Commits 13 Branches 8 Tags
4e784a4f4e95eb8a1a84ff185bdbd92e8a1ddde2
Commit Graph
7 Commits
Author SHA1 Message Date
slahiriandClaude Opus 4.5 365922dc67 Add workflow screenshot to README 
🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-09 01:46:53 +05:30
slahiriandClaude Opus 4.5 cd2a88c03d Add comprehensive NSFW body/clothing attributes to Generator UI design 
- Add content_detail parameter with 4 levels: minimal, standard, detailed, explicit
- Add detailed body_upper attributes: bust size/shape/cleavage, torso, midriff
- Add full body_exposure mapping: chest, stomach, back, hips, legs, buttocks
- Add extended clothing attributes: coverage, cutouts, swimwear styles, lingerie
- Add sensuality indicators in pose attributes
- Add YAML examples for each content detail level
- Document how explicit level captures full NSFW detail for Z-Image

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 20:46:05 +05:30
slahiriandClaude Opus 4.5 b47efcd652 Add seed control, prompt modes, and focus area toggles 
Seed Control:
- seed_mode: fixed, randomize, increment, decrement
- Fixed mode: deterministic caching (same input = same output)
- Randomize: fresh seed each run
- Inc/Dec: sequential variation generation

Prompt Direction:
- user_prompt: guide or override the analysis
- prompt_mode: analyze, enhance, priority, override
  * analyze: pure image analysis (ignore prompt)
  * enhance: prompt guides analysis (default)
  * priority: prompt first, image fills gaps
  * override: prompt dominates, minimal image reference

Focus Area Toggles:
- focus_subject: detailed subject description
- focus_environment: background/setting details
- focus_lighting: light source and quality
- focus_colors: materials and textures
- focus_mood: atmosphere and emotion
- include_text_quotes: quote visible text

Token Budget Allocation based on enabled toggles

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 20:39:23 +05:30
slahiriandClaude Opus 4.5 9e1df41851 Comprehensive agentic UI design with 56+ photography genres 
Scene Classification:
- 11 shot framing types (ECU to VLS)
- 56 photography genres across 6 categories:
  * People (13): Portrait, Headshot, Fashion, Beauty, Glamour, etc.
  * Event (7): Wedding, Concert, Sports, Street, etc.
  * Nature (15): Landscape, Wildlife, Macro, Astro, etc.
  * Commercial (7): Product, Food, Automotive, etc.
  * Artistic (10): Fine Art, Abstract, Surreal, etc.
  * Lifestyle (4): Travel, Candid, Selfie, etc.

Agentic Pipeline:
- Stage 1: Classification (shot + genre + tags)
- Stage 2: Metadata extraction (dimensions, aspect)
- Stage 3: Attribute mapping (rule-based per scene)
- Stage 4: Detailed LLM analysis
- Stage 5: Prompt composition with token tracking
- Stage 6: Z-Image recommendations

Outputs: prompt, structured_data, metadata, debug_log

Sources: Adobe, Pixpa, Shutterstock, StudioBinder, MasterClass

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 20:30:34 +05:30
slahiriandClaude Opus 4.5 03d9fdee79 Add UI design specification for SID_Z_Image_Prompt_Generator 
Visual wireframe and complete specification including:
- 12 input parameters organized in 5 UI sections
- Detail levels: Minimal, Detailed, Technical, Comprehensive
- 8 color styles, 20 photographer styles, 31 lighting options
- Widget specs: sliders, combos, multiline text, seed control
- Tooltips for all parameters
- Comparison with existing nodes
- Example usage flow with sample output

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 20:02:36 +05:30
slahiriandClaude Opus 4.5 86238c87d4 Update integration plan with two-node architecture 
Split into two separate nodes:
- SID_Z_Image_Prompt_Generator: Image → Z-Image prompt (vision-based)
- SID_Z_Image_Prompt_Enhancer: Simple text → Enhanced Z-Image prompt

Key changes:
- Generator uses Claude vision to analyze images
- Enhancer transforms simple prompts without requiring images
- Shared utilities in utils/zimage_utils.py
- Subject-specific guidance (Portrait, Landscape, Product, Action, Architecture)
- Enhancement levels (Minimal, Detailed, Comprehensive)
- Existing SID_AIPromptGenerator remains unchanged

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 19:46:44 +05:30
slahiriandClaude Opus 4.5 6ea645a0d2 Add Z-Image integration research and implementation plan 
- Z-IMAGE-PROMPTING-GUIDE.md: Comprehensive best practices for Z-Image prompts
  - Core principles (camera-first approach, visible descriptions only)
  - Technical settings (guidance_scale=0.0, no negative prompts)
  - Prompt structure and visual vocabulary
  - Genre-specific formulas (portrait, fashion, action, product)
  - Text quotation syntax for text-in-image
  - Anti-patterns to avoid

- Z-IMAGE-INTEGRATION-PLAN.md: Implementation plan for toolkit enhancement
  - New parameters: output_format, text_quote_style, max_narrative_length
  - New output: narrative_prompt (Z-Image compatible)
  - System prompt template for narrative generation
  - Output cleaning functions
  - Testing plan

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-08 19:37:54 +05:30