nodes/Limbicnation-ComfyUI-PromptGenerator
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
36 Commits 4 Branches 6 Tags
8af678564e89b76a3b43e6e20425a1249830ea0a
Commit Graph
36 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Gero Doll 8af678564e Merge pull request #5 from Limbicnation/feature/style-system-consolidation 
Feature/style system consolidation
 2026-02-27 06:12:49 +01:00
limbicnation 0829dfd336 fix: address code review findings for style system consolidation 2026-02-27 06:10:03 +01:00
limbicnation 3f0ee9b3ee chore: remove unwanted files and update .gitignore 2026-02-27 05:50:50 +01:00
limbicnation 272eb01b6c feat: consolidate style system to 9 styles across both nodes 2026-02-27 05:33:48 +01:00
limbicnation 44ac55e950 docs: fix README icon reference and update heading layout 
Point image src to icon.png, remove emoji from heading,
add clear float for proper badge alignment.
 2026-02-16 06:32:38 +01:00
limbicnation 394f13c463 docs: update CLAUDE.md to reflect v1.1.6 project state 
Sync version, document streaming API with per-chunk timeouts,
add model discovery section, restructure LoRA docs.
 2026-02-16 06:13:21 +01:00
limbicnation 6845d8b95f Added .claude/ and AGENTS.md to .gitignore 2026-02-16 06:08:39 +01:00
Gero Doll 16dcc6526c Merge pull request #4 from Limbicnation/fix/streaming-timeout-v1.1.6 
fix: replace blocking ollama.generate() with streaming to eliminate 1…
 2026-02-16 06:05:02 +01:00
limbicnation 6e2688f5ba fix: replace blocking ollama.generate() with streaming to eliminate 120s timeout errors 
Switch to streamed generation with per-chunk timeout enforcement and
ComfyUI ProgressBar integration. Add user-configurable timeout slider
(30-600s), cold-start detection via ollama.ps(), and graceful fallback
to subprocess. Bump version to 1.1.6.
 2026-02-16 06:01:47 +01:00
limbicnation b171550164 fix: replace blocking ollama.generate() with streaming to eliminate 120s timeout errors 
Switch to streamed generation with per-chunk timeout enforcement and
ComfyUI ProgressBar integration. Add user-configurable timeout slider
(30-600s), cold-start detection via ollama.ps(), and graceful fallback
to subprocess. Bump version to 1.1.6.
 2026-02-16 05:49:15 +01:00
limbicnation 1ded90171a fix: add video_wan style and improve LoRA compatibility v1.1.5 
- Add minimal 'video_wan' style template optimized for WanVideo LoRA
- Fix issue where LoRA would return empty results for verbose templates
- Improve extract_final_prompt to remove 'None' artifacts
- Bump version to 1.1.5
 2026-02-02 05:27:27 +01:00
limbicnation 9ddae78e22 chore: bump version to 1.1.4 and update developer agent memory 2026-02-02 05:21:37 +01:00
limbicnation bee70be519 Update version -> 1.1.4 2026-02-02 04:55:31 +01:00
limbicnation 4a8e9ba3ec fix: use 'model' key for Ollama model extraction 2026-02-02 04:54:02 +01:00
limbicnation 8f1a1810f7 chore: resolve merge conflict in pyproject.toml 2026-02-02 04:43:21 +01:00
limbicnation ac9f2f1cbb feat: implement dynamic LoRA selection and model discovery v1.1.3 
- Dynamic Ollama model discovery with 60s caching
- Prioritize LoRA-enhanced models (keywords: lora, limbicnation, fine)
- Authoritative model identification using full name (e.g., qwen3:8b)
- Fixed UI regression where model dropdown was hidden
- Added Modelfile template and usage documentation
 2026-02-02 04:41:25 +01:00
Gero Doll 01f951d22b Merge pull request #2 from Limbicnation/feature/lora-integration 
Feature/lora integration
 2026-02-02 04:04:36 +01:00
limbicnation eb884bd664 Update version to 1.1.2 2026-02-02 04:04:23 +01:00
limbicnation 7426ef8602 fix: use authoritative model name field only 
Addresses code review feedback to avoid ambiguous model identifiers.
The 'model' field may contain only base names (e.g., 'qwen3') while
'name' contains the full identifier ('qwen3:8b'). Skipping models
without proper 'name' field is safer than using potentially incorrect
fallback data.
 2026-02-02 04:03:14 +01:00
limbicnation 1455268c32 docs: update project memory with LoRA integration milestones 2026-02-02 03:56:55 +01:00
limbicnation 3f600fdf10 feat: implement dynamic LoRA model selection v1.1.1 2026-02-02 03:54:04 +01:00
Gero Doll d3f94dfd3f Merge pull request #1 from Limbicnation/feature/style-applier-node 
Feature/style applier node
 2026-01-28 02:13:52 +01:00
limbicnation 45cebe95b4 feat(nodes): add StyleApplierNode and still_image style (v1.0.5) 
- Add StyleApplierNode for Cinematic/Still Image style keywords
- Add style_presets.py with keyword-based style system
- Add still_image to PromptGeneratorNode dropdown and templates
- Update templates.yaml with still_image template
- Bump version to 1.0.5
 2026-01-28 02:13:14 +01:00
limbicnation aa8abd8d3f docs: add Publishing section with v1.0.4 registry info and deployment commands 2025-12-14 18:16:01 +01:00
limbicnation 5267a11165 chore: bump version to 1.0.4 for registry release 
- Update pyproject.toml version
- Add GitHub Actions publish workflow
- Update README
- Remove unused prompt-gen-icon.png
v1.0.4 		2025-12-14 18:10:18 +01:00
limbicnation 334aad5d4b docs: add CLAUDE.md with project overview and development guidelines 2025-12-14 07:22:16 +01:00
limbicnation 9066a4c6cf bump version = 1.0.3 2025-12-13 02:36:03 +01:00
limbicnation b85b3f6d95 fix: Use full URL for icon, bump to v1.0.2 2025-12-13 02:35:20 +01:00
limbicnation 812b847da9 bump version = 1.0.2 2025-12-13 02:31:29 +01:00
limbicnation 806346cf7d renamed icon 2025-12-13 02:30:53 +01:00
limbicnation 7b313b8d76 fix: Resize icon to 256x256, add to README, bump version to 1.0.1 2025-12-13 02:27:37 +01:00
limbicnation d75163f62b fix: Use correct license format for registry 2025-12-13 02:19:59 +01:00
limbicnation bacc675e72 add prompt-gen icon 2025-12-13 02:04:17 +01:00
limbicnation d03d9577c0 docs: Add feature image, workflow, and Gradio app link 2025-12-13 00:51:31 +01:00
limbicnation a2431e37a7 feat: Add ComfyUI Prompt Generator node 
- PromptGeneratorNode with 7 style presets (cinematic, anime, photorealistic, fantasy, abstract, cyberpunk, sci-fi)
- Temperature and Top-P sampling controls via Ollama API
- Optional emphasis (focus area) and mood inputs
- Reasoning toggle to show/hide model's thinking process
- Jinja2 templating for customizable style templates
- Graceful fallback when optional dependencies unavailable
- ComfyUI Registry ready (pyproject.toml configured)
 2025-12-13 00:11:52 +01:00
Gero Doll 642ebb67e1 Initial commit 2025-12-12 23:59:31 +01:00