Compare commits

..

204 Commits

Author SHA1 Message Date
vegu-ai-tools
9722936af6 free up space before container build 2025-12-05 20:53:57 +02:00
vegu-ai-tools
ca1e49075a fix ffmpeg verify 2025-12-05 20:53:46 +02:00
vegu-ai-tools
3f6ec9328c attempt to free space for container build 2025-12-05 20:31:41 +02:00
vegu-ai-tools
7bd8da47fc fix comment 2025-12-05 20:15:00 +02:00
vegu-ai-tools
c036aeb5eb Enhance Dockerfile comments for FFmpeg installation process, clarifying the choice of BtbN builds and the reliability of the direct download approach. 2025-12-05 20:14:01 +02:00
vegu-ai-tools
e06ded8ff4 Update Dockerfile to include FFmpeg 8.0 installation with shared libraries and set LD_LIBRARY_PATH for runtime compatibility. 2025-12-05 19:59:41 +02:00
vegu-ai-tools
974786a2ef remove comment 2025-12-05 17:51:11 +02:00
vegu-ai-tools
19f30a5991 docs 2025-12-05 17:27:22 +02:00
vegu-ai-tools
82f72d2a0d Add SHA256 checksum verification for FFmpeg download in install-ffmpeg.bat to ensure file integrity. 2025-12-05 17:23:32 +02:00
vegu-ai-tools
735cdfee51 Add optional FFmpeg installation step in install.bat, providing warnings for failure or absence of the script. 2025-12-05 17:15:16 +02:00
vegu-ai-tools
868dd924d1 relock 2025-12-05 17:04:20 +02:00
vegu-ai-tools
bdbd028bc8 docs 2025-12-05 16:55:14 +02:00
vegu-ai-tools
61632464bd Update FFmpeg installation script to download version 8.0.1 from GitHub and use tar for extraction, simplifying the process and ensuring compatibility with Windows 10. 2025-12-05 16:53:02 +02:00
vegu-ai-tools
b10ff490bb Add FFmpeg installation script: create install-ffmpeg.bat to automate downloading, extracting, and verifying FFmpeg installation for Talemate. 2025-12-05 16:40:47 +02:00
vegu-ai-tools
20380e417b relock 2025-12-05 15:49:32 +02:00
vegu-ai-tools
f6353ced38 relock 2025-12-05 15:16:51 +02:00
vegu-ai-tools
7f6c8c7dd9 add torchcodec 2025-12-05 15:07:24 +02:00
vegu-ai-tools
acca2a0fcf linting 2025-11-30 14:18:10 +02:00
vegu-ai-tools
5d48cc06d4 fix chara cv3 spec support 2025-11-30 14:07:08 +02:00
vegu-ai-tools
3e206e5c06 linting 2025-11-29 21:27:53 +02:00
vegu-ai-tools
f455a11096 Add methods for caching model data in Backend class: implement _get_cache_data and _apply_cache_data to enhance data sharing between backend instances. 2025-11-29 21:24:07 +02:00
vegu-ai-tools
647e2f277c Add caching functionality to Backend class: introduce methods for retrieving and applying cached data after successful test connections, enhancing data management across backend instances. 2025-11-29 21:24:02 +02:00
vegu-ai-tools
8cbecc67f2 Enhance sdnext_emit_status method to update model choices based on active handlers, improving backend responsiveness during processing. 2025-11-29 21:16:04 +02:00
vegu-ai-tools
91a7df67b1 Add model management in Backend class: initialize models list, update on_status_change to refresh model choices, and modify sdnext_update_model_choices to accept backend parameter for improved flexibility. 2025-11-29 21:08:00 +02:00
vegu-ai-tools
1d032e3309 Implement updateChoicesOnly method in AgentModal to preserve unsaved user values when updating agent choices. Update AIAgent to call this method when the modal is open and the current agent is being updated. 2025-11-29 21:07:50 +02:00
vegu-ai-tools
87089fad46 Add on_status_change method to Backend class for status updates, triggering visual agent notifications on status changes. 2025-11-29 21:07:29 +02:00
vegu-ai-tools
206939559c Update status icon fallback in VisualLibrary component to use 'mdi-minus-circle' for undefined statuses, improving visual feedback for users. 2025-11-29 19:05:42 +02:00
vegu-ai-tools
1757070bb9 Refactor API URL handling in visual backends to use a new utility function for normalization, ensuring consistent URL formatting across automatic1111, comfyui, and sdnext backends. 2025-11-29 19:03:09 +02:00
vegu-ai-tools
49689bcf00 Add visual and voice library documentation 2025-11-29 18:04:25 +02:00
vegu-ai-tools
c113a34a35 docs 2025-11-29 18:02:31 +02:00
vegu-ai-tools
6c5b95d501 docs 2025-11-29 17:52:34 +02:00
vegu-ai-tools
532a0a05b7 docs 2025-11-29 16:47:14 +02:00
vegu-ai-tools
04ec39f154 visual library docs 2025-11-29 15:14:24 +02:00
vegu-ai-tools
a5368a8a66 Update documentation for Shared Context to Shared World, enhancing clarity and adding details on episodes management and scene linking. 2025-11-29 13:57:33 +02:00
vegu-ai-tools
8ffdd8841c docs 2025-11-29 13:38:40 +02:00
vegu-ai-tools
83e2bdbe0e character card import docs 2025-11-29 13:26:44 +02:00
vegu-ai-tools
c97a83fb9a docs images 2025-11-29 13:07:51 +02:00
vegu-ai-tools
e2a1e465dd dont give saved character card image files as an option for character import 2025-11-29 13:07:42 +02:00
vegu-ai-tools
639955edaf Fix model name cleaning function to handle backslashes in template file names 2025-11-29 02:46:22 +02:00
vegu-ai-tools
366efb2532 linting 2025-11-28 21:36:13 +02:00
vegu-ai-tools
0cb4ba17ef add reason_prefill to force model to start with thinking block 2025-11-28 21:36:02 +02:00
vegu-ai-tools
f3b398322b linting 2025-11-28 19:56:03 +02:00
vegu-ai-tools
5e46107d07 fix kcpp visual agent auto setup failure messages 2025-11-28 19:54:34 +02:00
vegu-ai-tools
c8162ad350 linting 2025-11-28 15:57:49 +02:00
vegu-ai-tools
590e33c9b0 improve backend status display in visual library 2025-11-28 15:57:19 +02:00
vegu-ai-tools
5860745091 Remove custom clear tags functionality from VisualImageView component 2025-11-28 15:35:32 +02:00
vegu-ai-tools
4e2ad01f81 Add clear tags functionality in VisualImageView component
- Introduced a new inline confirmation button to clear tags from the form.
- Enhanced the component by importing and integrating the ConfirmActionInline component.
- Added logic to reset the tags array when the clear action is confirmed.
2025-11-28 15:28:42 +02:00
vegu-ai-tools
66bbe267e4 Add tag management features in VisualImageView component
- Introduced copy and paste functionality for tags using clipboard operations.
- Added tooltips for copy and paste buttons to enhance user experience.
- Improved layout of action buttons for better alignment and accessibility.
2025-11-28 15:25:00 +02:00
vegu-ai-tools
fcd675def0 Enhance asset search functionality in VisualLibraryScene component
- Added clear button to asset search input for easier user interaction.
- Implemented logic to determine which folders to open based on the current asset search input, ensuring relevant folders are displayed when filtering is active.
- Merged computed folder IDs with currently open nodes to maintain user selections during searches.
2025-11-28 15:20:51 +02:00
vegu-ai-tools
32dfc21c2f linting 2025-11-28 14:19:04 +02:00
vegu-ai-tools
fdf5ec0426 min thinking tokens to 512 2025-11-28 14:00:48 +02:00
vegu-ai-tools
598f6ad0b3 z image turbo workflow 2025-11-28 13:55:54 +02:00
vegu-ai-tools
b92b1048ae Add workflow reloading functionality to Backend class and improve status checks
- Implemented `_reload_workflow_if_outdated` method to reload the workflow from disk if it is outdated.
- Updated `ready` method to call the new workflow reloading method and added additional status checks for workflow validity.
- Enhanced logging for various workflow status conditions to improve debugging and monitoring.
2025-11-28 13:46:02 +02:00
vegu-ai-tools
27c378128f linting 2025-11-25 03:20:07 +02:00
vegu-ai-tools
61c0307543 sdnext auth method 2025-11-25 03:19:40 +02:00
vegu-ai-tools
8ab47e751e Improve debug logging in asset path retrieval by clarifying log message when asset ID is not found. 2025-11-25 02:46:37 +02:00
vegu-ai-tools
b636cb5398 nodes 2025-11-23 02:17:25 +02:00
vegu-ai-tools
6986f4df3f fix issue when attempting to load scene from scene search list 2025-11-23 02:15:47 +02:00
vegu-ai-tools
8ec0188001 Enhance name formatting in WorldState class to correctly handle possessive cases by using regex, ensuring proper title casing for character and item names. 2025-11-22 21:04:23 +02:00
vegu-ai-tools
cb38f6a830 Update visual styles instructions in YAML template to emphasize clear artist guidance for various art styles. 2025-11-22 20:40:04 +02:00
vegu-ai-tools
65d9137b66 Refactor prompt building logic in VisualPrompt class for improved readability and maintainability. Adjust formatting in validation module to ensure consistent newline handling. 2025-11-22 20:00:15 +02:00
vegu-ai-tools
e8fc01d227 modules 2025-11-22 19:03:45 +02:00
vegu-ai-tools
a523ba8828 Add ValidateAssetID node for asset ID validation in validation module, enhancing asset management capabilities. 2025-11-22 19:02:50 +02:00
vegu-ai-tools
ca8c2d0e03 Remove unused mediaType property and related code from DirectorConsoleChatMessageAssetView component, streamlining image handling logic and improving code clarity. 2025-11-22 18:22:44 +02:00
vegu-ai-tools
8ed3c9badf Refactor image preview functionality in DirectorConsoleChatMessageAssetView component. Replace dialog trigger with a dedicated method to handle image loading and preview sizing based on image dimensions. 2025-11-22 18:18:59 +02:00
vegu-ai-tools
0d7d49f107 nodes 2025-11-22 17:29:34 +02:00
vegu-ai-tools
8bcec0acb8 Add create_placeholder property to ValidateCharacter for optional placeholder creation when character does not exist, enhancing validation logic. 2025-11-22 17:28:52 +02:00
vegu-ai-tools
dc3ba4a18a Add new properties to VisualPrompt for handling positive and negative prompts, including keywords and descriptive formats, and refactor prompt building logic. 2025-11-22 17:28:42 +02:00
vegu-ai-tools
16f135c8f0 linting 2025-11-22 16:16:44 +02:00
veguAI
2d6d61d9d6 Director visuals (#22)
director chat action for image creation
2025-11-22 16:16:06 +02:00
vegu-ai-tools
19b70c4ba8 missing test files 2025-11-21 00:28:31 +02:00
vegu-ai-tools
e6fbc2dfaf updated what's new 2025-11-20 23:41:45 +02:00
vegu-ai-tools
053a872fde Enhance delete confirmation flow in WorldStateManagerCharacter component by focusing on input field after clicking delete button. 2025-11-20 22:53:34 +02:00
vegu-ai-tools
c0e224e2b6 linting 2025-11-20 22:29:20 +02:00
vegu-ai-tools
7fc2fccd0c Add gemini-3-pro-image-preview option to GoogleImageMixin for enhanced model selection. 2025-11-20 22:29:10 +02:00
vegu-ai-tools
7c253650b6 Add first_token_time to RequestInformation for accurate rate calculation 2025-11-20 19:44:56 +02:00
vegu-ai-tools
11c1427d9a Reduce debounce time for scene status emission from 50ms to 25ms for improved responsiveness. 2025-11-20 19:43:53 +02:00
vegu-ai-tools
7e376e02b0 node update 2025-11-20 14:01:34 +02:00
vegu-ai-tools
bf326b636d debounce scene emit_status 2025-11-20 13:32:57 +02:00
vegu-ai-tools
8502306824 Replace v-text-field with v-number-input in multiple components for improved number input handling and consistency. 2025-11-20 13:21:52 +02:00
vegu-ai-tools
8a16f500df Refactor import options in CharacterCardImport component to use v-checkbox with tooltips for better user experience and clarity. 2025-11-20 12:56:39 +02:00
vegu-ai-tools
18f8c4f752 json character card import should use new system 2025-11-20 12:39:41 +02:00
vegu-ai-tools
643ca7fc32 cleanup 2025-11-20 12:11:36 +02:00
vegu-ai-tools
1eba909376 cleanup 2025-11-20 12:02:00 +02:00
vegu-ai-tools
ac3e569324 linting 2025-11-20 11:51:41 +02:00
vegu-ai-tools
e632f2bff6 Add unique random color assignment for characters in load_scene_from_character_card function 2025-11-20 11:50:25 +02:00
vegu-ai-tools
dde6c76aa9 Add writing style template selection to CharacterCardImport component and update scene loading logic to apply selected template 2025-11-20 11:41:26 +02:00
vegu-ai-tools
152f47a12e Update episode management to reflect correct directory structure for episodes.json 2025-11-20 02:09:43 +02:00
vegu-ai-tools
a2f12ffb1b Add maxHeight property to v-select in AgentModal component 2025-11-20 02:01:54 +02:00
vegu-ai-tools
b580eb2316 Enhance error logging in Google Image backend to include result data on parse errors 2025-11-20 02:00:01 +02:00
vegu-ai-tools
7acb26cf55 fix asset meta when created through auto cover image creation 2025-11-20 01:44:54 +02:00
vegu-ai-tools
4fd3e0d25e when introducing a character without instructions send empty content 2025-11-20 00:29:41 +02:00
vegu-ai-tools
495d9b1abe prompt tweaks 2025-11-20 00:29:23 +02:00
vegu-ai-tools
2ad54cc5f2 Update cover image IDs in generate_intro_scenes 2025-11-20 00:10:38 +02:00
vegu-ai-tools
31c4e20612 Add 'gemini-3-pro-preview' model option to GoogleImageMixin configuration 2025-11-19 12:15:18 +02:00
vegu-ai-tools
63db597c1c Add support for new model 'gemini-3-pro-preview' in Google client 2025-11-19 12:09:01 +02:00
vegu-ai-tools
6fefefc695 linting 2025-11-19 01:44:19 +02:00
vegu-ai-tools
7391d76dc2 Add character dialogue example generation functionality
- Implemented a new method in the CharacterCreatorMixin to extract or generate dialogue examples for characters based on provided text.
- Updated the character card loading process to include dialogue example determination, ensuring examples are regenerated properly.
- Created a new Jinja2 template for generating dialogue examples, including guidelines for format and content.
- Enhanced logging for dialogue example generation to track character names and example counts.
2025-11-19 01:36:16 +02:00
vegu-ai-tools
885b48a83f Add new Jinja2 template for generating horizontal scene illustrations
- Created a new template to generate prompts for horizontal illustrations capturing dynamic moments in scenes.
- Included sections for character context, requirements, and task instructions to guide image generation.
- Emphasized the importance of action, emotion, and cinematic framing in the generated images.
2025-11-18 20:34:13 +02:00
vegu-ai-tools
b364bc28b0 Fix example stripping logic in Character class to handle cases without colons 2025-11-18 20:20:47 +02:00
vegu-ai-tools
2e43da8b0f linting 2025-11-18 20:09:01 +02:00
vegu-ai-tools
0e393a09e2 Add require_active flag to reinforcement templates and logic
- Introduced a require_active boolean flag in various reinforcement classes and templates to control reinforcement activation based on character status.
- Updated logic in WorldStateAgent to skip inactive character reinforcements when require_active is true.
- Enhanced frontend components to support the new require_active option for character reinforcements.
2025-11-18 19:51:34 +02:00
vegu-ai-tools
a595e73c1e Update response length calculation in NarratorAgent to use max_generation_length 2025-11-18 19:35:33 +02:00
vegu-ai-tools
3c25b99340 Enhance character card import UI with re-analyze and retry options
- Added a "Re-analyze" button that appears when file data is present and analysis is not in progress.
- Improved error handling by introducing a "Retry" button in the error alert for failed analyses.
- Adjusted the display logic for analysis information to improve user experience.
2025-11-18 19:32:10 +02:00
vegu-ai-tools
fe014755e2 allow cancelling of character card import 2025-11-18 19:25:10 +02:00
vegu-ai-tools
35b11156ee fix character card preview when card is loaded from talemate file system 2025-11-18 13:51:59 +02:00
vegu-ai-tools
b733774096 fix shared world auto refresh 2025-11-18 13:41:37 +02:00
vegu-ai-tools
a89f11cd0a linting 2025-11-18 13:09:48 +02:00
vegu-ai-tools
33cc6d3442 Add max-height constraints to episode layouts and sidebars for improved UI consistency 2025-11-18 13:09:41 +02:00
vegu-ai-tools
79d39e5297 Remove max-height constraint from episodes layout to improve UI flexibility 2025-11-18 13:06:43 +02:00
vegu-ai-tools
70cd6946eb Update excluded character names to enhance detection accuracy
- Expanded the list of excluded names in character detection to include additional variations, preventing false positives during character name identification.
2025-11-18 13:04:44 +02:00
vegu-ai-tools
eb55ec6877 linting 2025-11-18 12:53:45 +02:00
vegu-ai-tools
218a301909 Remove early return in handleMessage to ensure prompts are cleared when scene is loaded 2025-11-18 12:52:39 +02:00
vegu-ai-tools
00b96b603f prompt tweaks 2025-11-18 12:52:25 +02:00
vegu-ai-tools
d663ee01bd Refactor import options initialization in load_scene_from_character_card function
- Moved the initialization of import_options to ensure it occurs after the import statements, improving code organization and readability.
2025-11-18 12:40:07 +02:00
vegu-ai-tools
fb28888011 dont pass agent instances around 2025-11-18 12:36:42 +02:00
vegu-ai-tools
186630b11c tweaks to character description and character sheet extraction 2025-11-18 12:32:23 +02:00
vegu-ai-tools
0f96b19af7 stuff 2025-11-18 05:14:43 +02:00
vegu-ai-tools
30377189c5 linting 2025-11-18 03:47:33 +02:00
vegu-ai-tools
13bf370e76 Enhance character detection by processing texts in chunks
- Introduced a new method, detect_characters_from_texts, to analyze multiple texts by processing them in manageable chunks based on the client's max context size.
- Added functionality to avoid duplicate detections by passing already detected character names.
- Implemented utility functions for chunking items by token count and removing substring names to improve character detection accuracy.
- Updated the corresponding Jinja2 template to reflect the changes in character detection logic.
2025-11-18 03:47:20 +02:00
vegu-ai-tools
d7df0dd2e5 linting 2025-11-18 03:18:58 +02:00
vegu-ai-tools
c5cb4d7335 Enhance character card validation and error handling
- Added validation to ensure character card data is a dictionary, raising a ValueError with a clear message if invalid.
- Improved error handling in the analyze_character_card function to provide user-friendly messages for various parsing errors, including JSON decoding and file not found issues.
- Updated the CharacterCardImport component to adjust dialog sizing based on analysis errors, enhancing user feedback during character import.
2025-11-18 03:18:36 +02:00
vegu-ai-tools
0db092cb8a linting 2025-11-18 03:10:27 +02:00
vegu-ai-tools
5496ef7a57 Enhance CharacterCardImport component with dynamic dialog sizing and image preview
- Updated the dialog max-width to adjust based on the analyzing state, improving user experience during character import.
- Added an image preview section that displays the character card image or relevant messages based on the file data and path, enhancing feedback during the import process.
2025-11-18 03:09:46 +02:00
vegu-ai-tools
4e999b8300 Enhance episode title generation and character import options
- Updated the CreatorAgent to parse titles from AI responses, ensuring titles are extracted from <TITLE> tags.
- Added a new option in CharacterCardImportOptions to enable episode title generation from alternate greetings.
- Implemented logic in the loading process to generate titles for episodes if the new option is enabled.
- Enhanced the UI in CharacterCardImport.vue to include a toggle for generating episode titles during character import.
2025-11-18 03:07:39 +02:00
vegu-ai-tools
cc8579b554 Refactor CharacterCardImport component to enhance character selection UI
- Replaced radio buttons with a tabbed interface for selecting character modes: Default Template, Detected Character, and Import from Another Scene.
- Updated the layout to use a window component for displaying the corresponding input fields based on the selected character mode.
- Improved overall user experience with better organization and visual clarity in character selection options.
2025-11-18 02:29:29 +02:00
vegu-ai-tools
5741157483 Enhance character parsing logic in character_card.py
- Updated _parse_characters_from_greeting_text to accept a scene parameter for improved character validation.
- Implemented logic to validate character names against actual characters in the scene, including partial matches for NPCs.
- Added functionality to activate up to 2 NPCs if no characters are detected from the greeting text.
- Adjusted related calls to ensure compatibility with the new parsing method.
2025-11-18 02:23:04 +02:00
vegu-ai-tools
50234a4d88 Refactor WorldStateManagerSceneSharedContext for improved layout and structure
- Simplified the template structure by removing unnecessary nested rows and columns in WorldStateManagerSceneSharedContext.vue.
- Enhanced the UI layout for better readability and user experience in managing shared contexts.
- Updated WorldStateManagerSceneSharedWorld.vue to maintain consistent structure when integrating the SharedContext component.
2025-11-18 02:11:06 +02:00
vegu-ai-tools
9ed511139f Refactor WorldStateManagerScene components and introduce SharedWorld
- Renamed WorldStateManagerSceneSharedContext to WorldStateManagerSceneSharedWorld for clarity.
- Created a new WorldStateManagerSceneSharedWorld component to encapsulate shared context functionality.
- Updated references and imports across components to reflect the new structure.
- Enhanced UI elements for better user experience in managing shared contexts and episodes.
2025-11-18 02:08:56 +02:00
vegu-ai-tools
3fe34c6688 Implement auto-selection of characters based on episode intro in WorldStateManagerSceneSharedContext
- Added functionality to automatically select characters mentioned in the episode intro when the dialog is open and characters are available.
- Introduced a new method, autoSelectCharactersFromIntro, to handle the selection logic based on character names found in the intro text.
- Enhanced the existing episode selection handling to trigger character auto-selection when an episode is selected.
2025-11-18 02:05:56 +02:00
vegu-ai-tools
85777a4d2a Add sharing functionality for world entries and characters in WorldStateManager
- Implemented methods to share and unshare all world entries and characters in the WorldStateManagerPlugin and CharacterMixin classes.
- Enhanced the WorldStateManagerSceneSharedContext.vue component with UI elements to trigger sharing actions for characters and world entries.
- Added corresponding websocket actions to facilitate sharing and unsharing operations from the frontend.
2025-11-18 01:59:36 +02:00
vegu-ai-tools
fe854141e7 Refactor WorldStateManagerScene components for improved UI and functionality
- Removed unnecessary icon slots in WorldStateManagerSceneEpisodes.vue for cleaner design.
- Updated icon colors in WorldStateManagerSceneSharedContext.vue to enhance visual hierarchy.
- Streamlined dialog components for creating new scenes and shared contexts, ensuring consistent layout and user experience.
2025-11-18 01:54:14 +02:00
vegu-ai-tools
da0be7a9ff Add shared context setup option in CharacterCardImport and related components
- Introduced a new toggle in CharacterCardImport.vue for setting up shared context during character import.
- Updated CharacterCardImportOptions model to include a flag for shared context setup.
- Implemented _setup_shared_context_for_import function in character_card.py to handle shared context creation and character marking.
- Enhanced WorldStateManagerSceneSharedContext.vue to display shared context details and manage linked files more effectively.
2025-11-18 01:51:19 +02:00
vegu-ai-tools
d78c2398e4 Update icons and styling in WorldStateManagerScene components for improved UI consistency
- Changed the delete button icon in WorldStateManagerSceneEpisodes.vue to mdi-close-circle-outline for better visual clarity.
- Added margin-top class to the card text in WorldStateManagerSceneSharedContext.vue to enhance layout spacing.
2025-11-18 01:30:14 +02:00
vegu-ai-tools
096af6486e Enhance WorldStateManagerScene components with new scene and shared context features
- Added a new section for creating scenes with shared context in WorldStateManagerSceneSharedContext.vue.
- Improved UI elements including buttons and chips for better user interaction.
- Updated episode action buttons in WorldStateManagerSceneEpisodes.vue for clarity and consistency.
2025-11-18 01:26:44 +02:00
vegu-ai-tools
8e3c6fe166 Add episode editing functionality and enhance WorldStateManagerScene components
- Introduced a dialog for adding and editing episodes in WorldStateManagerSceneEpisodes.vue.
- Added buttons for editing selected episodes and improved form validation.
- Updated WorldStateManagerScene and WorldStateManagerSceneSharedContext to accept templates and generation options as props for better context management.
2025-11-18 01:07:17 +02:00
vegu-ai-tools
b097fe9439 refactor scene_versions into episodes 2025-11-18 00:57:53 +02:00
vegu-ai-tools
b8c46996b2 move import 2025-11-17 01:34:01 +02:00
vegu-ai-tools
b3cdf11d9d linting 2025-11-17 01:32:17 +02:00
vegu-ai-tools
f238b890b4 Refactor scene loading by introducing scene_stub function to create minimal Scene objects, improving asset access without full scene loading. Update SceneAssets class to utilize scene_stub for better efficiency. 2025-11-17 01:26:34 +02:00
vegu-ai-tools
30ce1541b9 transfer cover image 2025-11-17 01:22:56 +02:00
vegu-ai-tools
a46bdb929c linting 2025-11-16 23:02:09 +02:00
vegu-ai-tools
b7663ab263 Implement player character setup options in CharacterCardImport component, including template creation, existing character selection, and scene import. Update scene loading logic to accommodate new character data format and enhance character management in the application. 2025-11-16 23:01:56 +02:00
vegu-ai-tools
584b3ae7b9 modules 2025-11-16 22:24:19 +02:00
vegu-ai-tools
dd6181d02a Rename "Restore Scene" to "Reset" in SceneToolsSave component for clarity, and update confirmation prompt action label to "Reset Scene". 2025-11-16 21:31:30 +02:00
vegu-ai-tools
a2c97932d7 Enhance scene restoration process by adding a confirmation prompt in SceneToolsSave component and ensuring restore_from state is preserved in TaleMate class. 2025-11-16 21:25:57 +02:00
vegu-ai-tools
267b552d38 Add save and restore functionality to SceneToolsSave component, including visual indicators for unsaved changes and the ability to restore scenes from specified points. 2025-11-16 21:19:04 +02:00
vegu-ai-tools
814d006c88 Add unified API key support across various components and implement ConfigWidgetUnifiedApiKey for streamlined API key management. 2025-11-16 18:47:00 +02:00
vegu-ai-tools
4e57761553 Refactor CoverImage component by removing unused isPortrait logic and simplifying image handling, enhancing code clarity and maintainability. 2025-11-16 18:28:33 +02:00
vegu-ai-tools
389f0ee9ad linting 2025-11-16 18:25:24 +02:00
vegu-ai-tools
6a5d1f1173 Refactor VisualImageView component to use v-card for image display and adjust image preview styles for better responsiveness 2025-11-16 18:25:14 +02:00
vegu-ai-tools
da739b1d20 move buttons to scene component 2025-11-16 18:10:37 +02:00
vegu-ai-tools
2116d6e552 assets 2025-11-16 18:10:19 +02:00
vegu-ai-tools
36b6ef5b7d add analysis data 2025-11-16 17:40:21 +02:00
vegu-ai-tools
3d4d995710 Add reference_assets field to EditAssetMetaPayload and update related components for improved asset management and metadata handling. 2025-11-16 17:40:12 +02:00
vegu-ai-tools
13087e676a Add metadata to scene cover asset in library.json, including visual type, generation type, resolution, and tags for improved asset management. 2025-11-16 17:26:15 +02:00
vegu-ai-tools
3575a0d67b assets 2025-11-16 17:24:43 +02:00
vegu-ai-tools
b28c28f25d Refine template management description in Templates.vue for clarity and conciseness 2025-11-16 17:23:43 +02:00
vegu-ai-tools
1e9b8f18a1 Add styles for image positioning in IntroRecentScenes component to enhance visual layout 2025-11-16 16:46:56 +02:00
vegu-ai-tools
53dbc2a085 Remove redundant debug logging in migrate_scene_assets_to_library function 2025-11-16 16:46:51 +02:00
vegu-ai-tools
5a12395fb9 assets 2025-11-16 16:41:17 +02:00
vegu-ai-tools
df197ac873 assets 2025-11-16 16:40:59 +02:00
vegu-ai-tools
deab1bf3c3 assets 2025-11-16 16:39:28 +02:00
vegu-ai-tools
0f5731f7b5 Add scene_info method to SceneAssets class for simplified asset data retrieval 2025-11-16 16:24:22 +02:00
vegu-ai-tools
5b5a9e3d68 update assets 2025-11-16 16:24:00 +02:00
vegu-ai-tools
5fed0115f3 Improve image aspect ratio handling in CoverImage component by introducing a delay before checking the aspect ratio. Reset portrait flag upon asset updates for various asset types to ensure accurate orientation detection. 2025-11-16 16:10:51 +02:00
vegu-ai-tools
2589ad36f8 linting 2025-11-16 14:45:52 +02:00
vegu-ai-tools
d2441142e0 module update 2025-11-16 14:45:42 +02:00
vegu-ai-tools
2a645ee0c6 Implement save option for asset analysis in AnalysisRequest and related components. Update websocket handler and frontend to support saving analysis results conditionally based on user input. 2025-11-16 14:45:14 +02:00
vegu-ai-tools
d453ee6781 linting 2025-11-16 13:39:55 +02:00
vegu-ai-tools
eaa9f76181 unified asset library 2025-11-16 13:38:55 +02:00
vegu-ai-tools
6890f4f138 Add unified API key configuration support across various client classes and implement handling for saving the API key in the server config. Update frontend components to manage and display the unified API key. 2025-11-16 12:31:18 +02:00
vegu-ai-tools
1ae3d2669c Enhance CoverImage component with improved aspect ratio handling and responsive styling. Added logic to determine image orientation and updated class bindings for better visual consistency. 2025-11-16 11:44:05 +02:00
vegu-ai-tools
566e03c9e6 Refactor CoverImage and IntroRecentScenes components to improve image handling and styling. Added aspect ratios and updated class usage for better responsiveness and visual consistency. 2025-11-16 11:32:54 +02:00
vegu-ai-tools
dff778c997 relock 2025-11-16 11:28:06 +02:00
vegu-ai-tools
a8d9202c92 Add clarification on content search methodology in WorldStateManagerContextDB.vue 2025-11-16 02:00:15 +02:00
vegu-ai-tools
9dfe5d34db more character info for visual generation 2025-11-16 01:55:59 +02:00
vegu-ai-tools
70fc83afe5 linting 2025-11-16 01:29:03 +02:00
vegu-ai-tools
1730c7f53c director can generate visuals for new characters 2025-11-16 01:28:40 +02:00
vegu-ai-tools
4cdf3f240f linting 2025-11-15 21:07:56 +02:00
vegu-ai-tools
693052046b Enhance loading status setup in character card import by adding parameters for character count and character book presence. Adjust loading steps calculation for improved accuracy based on these parameters. 2025-11-15 21:07:10 +02:00
vegu-ai-tools
b7ea6c9d7f linting 2025-11-15 21:03:48 +02:00
vegu-ai-tools
77a14ab3db Refactor character management to utilize ClientContext for scene detection, enhancing flexibility in handling active scenes. Update context model to include 'requires_active_scene' attribute and streamline error handling in client base. 2025-11-15 21:03:39 +02:00
vegu-ai-tools
0a0cf9427a linting 2025-11-15 20:45:47 +02:00
vegu-ai-tools
170380becc Remove unused 'app-ready' prop from VisualLibrary component to streamline code and improve clarity. 2025-11-15 20:45:03 +02:00
vegu-ai-tools
8ff51def0f Update character card import alert color and icon for improved visibility during analysis 2025-11-15 20:44:56 +02:00
vegu-ai-tools
7431752afe improved character card import 2025-11-15 20:30:43 +02:00
vegu-ai-tools
e4ae19949a Refactor changelog handling to use a local variable for InMemoryChangelog, improving code clarity and maintainability. 2025-11-15 20:17:04 +02:00
vegu-ai-tools
42d2b574db character card import improvements 2025-11-15 19:55:36 +02:00
vegu-ai-tools
833dde378f linting 2025-11-15 19:02:19 +02:00
vegu-ai-tools
3b7df1eb3c support alternative intros 2025-11-15 18:25:41 +02:00
vegu-ai-tools
7f2fc4421c character card lorebook loading 2025-11-15 17:42:36 +02:00
vegu-ai-tools
1a06b70994 Refactor scene cover image handling to emit status only when the scene is active, ensuring proper event management. Update TalemateApp to include 'type' prop for CoverImage component, enhancing flexibility in rendering. 2025-11-15 17:42:21 +02:00
vegu-ai-tools
858b94856b load.py into load module 2025-11-15 16:16:18 +02:00
vegu-ai-tools
5a38958819 Enhance tab selection logic in TalemateApp to prefer 'home' tab when no scene is loaded, improving user experience during initial app interactions. 2025-11-15 16:00:33 +02:00
vegu-ai-tools
c0eaa00872 Update client configuration options to only include enabled clients in the agent's config options. 2025-11-15 15:58:21 +02:00
vegu-ai-tools
7052cedd3f app ready status 2025-11-15 15:05:16 +02:00
vegu-ai-tools
0c77004725 fix pydantic warnings 2025-11-15 13:48:42 +02:00
vegu-ai-tools
93c5c58c10 fix tests 2025-11-15 13:44:00 +02:00
vegu-ai-tools
e5e3091c9c remove runpod integration 2025-11-15 13:40:24 +02:00
vegu-ai-tools
be472b4b15 module fix 2025-11-15 13:34:06 +02:00
veguAI
04e8349975 Visual refactor 2 (#20)
- visual agent refactor
- visual library
- move template management out of world editor
2025-11-15 13:32:50 +02:00
vegu-ai-tools
1f47219b35 enabled needs to be passed to apply config so agents can act on it changing 2025-10-26 21:59:49 +02:00
vegu-ai-tools
e51d1dbcbc set 0.34 2025-10-26 20:14:42 +02:00
7 changed files with 11 additions and 50 deletions

View File

@@ -1,6 +1,6 @@
[project]
name = "talemate"
version = "0.34.1"
version = "0.34.0"
description = "AI-backed roleplay and narrative tools"
authors = [{name = "VeguAITools"}]
license = {text = "GNU Affero General Public License v3.0"}

View File

@@ -256,10 +256,6 @@ class OpenRouterClient(ClientBase):
def openrouter_api_key(self):
return self.config.openrouter.api_key
@property
def requires_reasoning_pattern(self) -> bool:
return False
@property
def supported_parameters(self):
return [
@@ -368,9 +364,6 @@ class OpenRouterClient(ClientBase):
payload = {
"model": self.model_name,
"messages": messages,
"reasoning": {
"max_tokens": self.validated_reason_tokens,
},
"stream": True,
**parameters,
}
@@ -383,7 +376,6 @@ class OpenRouterClient(ClientBase):
)
response_text = ""
reasoning_text = ""
buffer = ""
completion_tokens = 0
prompt_tokens = 0
@@ -418,21 +410,14 @@ class OpenRouterClient(ClientBase):
try:
data_obj = json.loads(data)
delta = data_obj["choices"][0]["delta"]
content = delta.get("content")
reasoning = delta.get("reasoning")
content = data_obj["choices"][0]["delta"].get(
"content"
)
usage = data_obj.get("usage", {})
completion_tokens += usage.get(
"completion_tokens", 0
)
prompt_tokens += usage.get("prompt_tokens", 0)
if reasoning:
reasoning_text += reasoning
self.update_request_tokens(
self.count_tokens(reasoning)
)
if content:
response_text += content
# Update tokens as content streams in
@@ -447,15 +432,6 @@ class OpenRouterClient(ClientBase):
response_content = response_text
self._returned_prompt_tokens = prompt_tokens
self._returned_response_tokens = completion_tokens
self._reasoning_response = reasoning_text
self.log.debug(
"generated response",
response=response_content[:128] + " ..."
if len(response_content) > 128
else response_content,
reasoning_length=len(reasoning_text),
)
return response_content

View File

@@ -1,3 +1,3 @@
__all__ = ["VERSION"]
VERSION = "0.34.1"
VERSION = "0.34.0"

View File

@@ -1,12 +1,12 @@
{
"name": "talemate_frontend",
"version": "0.34.1",
"version": "0.34.0",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "talemate_frontend",
"version": "0.34.1",
"version": "0.34.0",
"dependencies": {
"@codemirror/lang-json": "^6.0.1",
"@codemirror/lang-markdown": "^6.2.5",

View File

@@ -1,6 +1,6 @@
{
"name": "talemate_frontend",
"version": "0.34.1",
"version": "0.34.0",
"private": true,
"type": "module",
"scripts": {

View File

@@ -76,25 +76,10 @@ export default {
data() {
return {
expand: false,
selected: "0.34.1",
selected: "0.34.0",
whatsNew: [
{
version: '0.34.1',
items: [
{
title: "OpenRouter Reasoning Model Fixes",
description: "Fixed issues with OpenRouter reasoning models.",
items: [
"Fix empty responses from reasoning models",
"Fix reasoning token not found errors",
"Let OpenRouter handle reasoning token collection",
"Configure reasoning effort correctly"
]
}
]
},
{
version: '0.34.1',
version: '0.34.0',
items: [
{
title: "Visual Agent Refactor",

2
uv.lock generated
View File

@@ -5227,7 +5227,7 @@ wheels = [
[[package]]
name = "talemate"
version = "0.34.1"
version = "0.34.0"
source = { editable = "." }
dependencies = [
{ name = "aiofiles" },