mirror of
https://github.com/vegu-ai/talemate.git
synced 2026-09-02 12:09:31 +02:00
53ecfe50290f5ea166e34c19b0a7df823aa32885
8 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
veguAI
|
121d103bea |
Pi Bridge: fix Windows generations never finalizing (#127)
* pi_bridge: add TALEMATE_PI_BRIDGE_TRACE event-stream instrumentation for #126 * fix(pi_bridge): shut pi down via stdin EOF with tree-kill fallback so Windows generations finalize (#126) * pi_bridge: address PR #127 review - bound all cleanup waits, survive repeat cancellation, cancellation tests, trace docs * pi_bridge: address PR #127 round-2 review - incremental stderr buffer, unconditional drain disposal, shielded tree kill, annotations, test handshake * pi_bridge: address PR #127 round-3 review - catch taskkill spawn failure, partial-stderr regression test, comment placement |
||
|
veguAI
|
c12a82930e |
0.38.0 (#272)
- **Per-Scene Agent Overrides** — Override agent configuration for a single scene without touching your global config. - **Message Revision History** — Browse and continue from previous regenerations / revisions. - **Scene Perspective Overrides** — Set distinct narrative perspectives for the player, NPCs, and narrator. - **Durable World State Snapshot** — The world state snapshot now persists and updates entities across refreshes. - **World State Highlights** — Characters, objects and places marked by the current world state snapshot are now clickable in the recent message(s) for additional detail. Plus 30 improvements and 17 bug fixes |
||
|
veguAI
|
f5d41c04c8 |
0.37.0 (#267)
0.37.0 - **Director Planning** — Multi-step todo lists in director chat plus a Generate long progress action for multi-beat scene arcs. - **Auto Narration** — Unified auto-narration replacing the old Narrate after Dialogue toggle, with a chance slider and weighted action mix. - **LLM Prompt Templates Manager** — Dedicated UI tab for viewing, creating, editing, and deleting prompt templates. - **Character Folders** — Collapsible folders in the World Editor character list, synced across linked scenes. - **OpenAI Compatible TTS** — Connect any number of OpenAI-compatible TTS servers in parallel. - **KoboldCpp TTS Auto-Setup** — KoboldCpp clients with a TTS model loaded register themselves as a TTS backend. - **Model Testing Harness** — Bundled scene that runs basic capability tests against any connected LLM. Plus 27 improvements and 28 bug fixes |
||
|
veguAI
|
42a8863e65 |
0.36.0 (#255)
Major Features - API key encryption at rest using Fernet (OS keyring with file fallback) - Prompt Manager: unified UI with template groups, priority ordering, override tracking, response extractors - Scene context history review panel with token budgets and best-fit mode - Multiple concurrent director chats with auto-generated titles - Granular scene state reset dialog - Time passage insert/edit/delete in scene view - Image analysis via OpenAI-compatible and Talemate Client backends - Volatile context placement after scene history for improved prompt caching Improvements - Configurable narrator generation length per narration type - AI Aware conversation mode - Summarizer: custom instructions, writing style inclusion, short line filtering - Anthropic: adaptive thinking support, updated model list (opus-4-5/4-6, haiku-4-5) - Google: gemini-3.1 support - World editor: generate from topic, quick create state reinforcement, reorganized menus - Node editor: promote scene modules to global - Frontend: version mismatch detection, hideable bracket content, required scene name - TTS: improved pause handling, audio tag support for vocal markers (ElevenLabs v3) - Writing style template for AI-generated instructions - Added Kimi.jinja2 LLM prompt template - Option to disable character names in stopping strings - Client response length enforcement options - Graduated token count sliders - Increased summarizer token threshold max Bugfixes - Fix bracket/paren/brace terminators stripped from message ends - Fix colon in conversation causing content loss - Fix "Use as reference" navigating to blank page - Fix avatar regeneration and manual regenerate - Fix conversation agent ignoring generation length - Fix duplicate length instructions with reasoning enabled - Fix trailing newline on message edits - Fix summarize dialogue sending too much context with layered history - Fix layered history inspection and construction issues - Fix empty response handling in summarization - Fix context ID dot notation with dotted character names - Fix recursive retry in focal agent - Fix leading whitespace causing duplicate prepared responses - Fix summarization not stripping ANALYSIS OF lines - Fix template group selection/removal in prompt manager - Fix multiline text in parentheses/brackets parser - Fix determine_character_name resolution - Fix character activate/deactivate desyncing creative menu - Fix character image generation missing context - Fix LMStudio client not sending token limits - Fix Recent Scene images on newer Chromium - Fix sequential reinforcement messages cut off at first linebreak - Fix reinforcement removal not clearing state - Fixes #252, #256, #258 Deprecations - Removed context investigations (replaced by AI-assisted RAG mixin) - Removed deprecated prompt templates (fix-continuity-errors, fix-exposition, etc.) - Removed conversation/edit.jinja2, auto break repetition, CLI reset layered history --------- Co-authored-by: theDTV2 <47825738+theDTV2@users.noreply.github.com> |
||
|
veguAI
|
d0ebe95ca6 |
0.35.0 (#242)
Major features: - Autonomous scene direction via director agent (replaces auto-direct) - Inline image display in scene feed - Character visuals tab for portrait/cover image management - Character message avatars with dynamic portrait selection - Pocket TTS and llama.cpp client support - Message appearance overhaul with configurable markdown display Improvements: - KoboldCpp: adaptive-p, min-p, presence/frequency penalty support - Setup wizard for initial configuration - Director chat action toggles - Visual agent: resolution presets, prompt revision, auto-analysis - Experimental concurrent requests for hosted LLM clients - Node editor alignment shortcuts (X/Y) and color picker Bugfixes: - Empty response retry loop - Client system prompt display - Character detail pins loading - ComfyUI workflow charset encoding - Various layout and state issues Breaking: Removed INSTRUCTOR embeddings |
||
|
veguAI
|
ce4c302d73 |
0.32.0 (#208)
* separate other tts apis and improve chunking * move old tts config to voice agent config and implement config widget ux elements for table editing * elevenlabs updated to use their client and expose model selection * linting * separate character class into character.pt and start on voice routing * linting * tts hot swapping and chunking improvements * linting * add support for piper-tts * update gitignore * linting * support google tts fix issue where quick_toggle agent config didnt work on standard config items * linting * only show agent quick toggles if the agent is enabled * change elevenlabs to use a locally maintained voice list * tts generate before / after events * voice library refactor * linting * update openai model and voices * tweak configs * voice library ux * linting * add support for kokoro tts * fix add / remove voice * voice library tags * linting * linting * tts api status * api infos and add more kokoro voices * allow voice testing before saving a new voice * tweaks to voice library ux and some api info text * linting * voice mixer * polish * voice files go into /tts instead of templates/voice * change default narrator voice * xtts confirmation note * character voice select * koboldai format template * polish * skip empty chunks * change default voice * replace em-dash with normal dash * adjust limit * replace libebreaks * chunk cleanup for whitespace * info updated * remove invalid endif tag * sort voices by ready api * Character hashable type * clarify set_simulated_environment use to avoid unwanted character deactivated * allow manual generation of tts and fix assorted issues with tts * tts websocket handler router renamed * voice mixer: when there are only 2 voices auto adjust the other weight as needed * separate persist character functions into own mixin * auto assign voices * fix chara load and auto assign voice during chara load * smart speaker separation * tts speaker separation config * generate tts for intro text * fix prompting issues with anthropic, google and openrouter clients * decensor flag off again * only to ai assisted voice markup on narrator messages * openrouter provider configuration * linting * improved sound controls * add support for chatterbox * fix info * chatterbox dependencies * remove piper and xtts2 * linting * voice params * linting * tts model overrides and move tts info to tab * reorg toolbar * allow overriding of test text * more tts fixes, apply intensity, chatterbox voices * confirm voice delete * lintinG * groq updates * reorg decorators * tts fixes * cancelable audio queue * voice library uploads * scene voice library * Config refactor (#13) * config refactor progres * config nuke continues * fix system prompts * linting * client fun * client config refactor * fix kcpp auto embedding selection * linting * fix proxy config * remove cruft * fix remaining client bugs from config refactor always use get_config(), dont keep an instance reference * support for reasoning models * more reasoning tweaks * only allow one frontend to connect at a time * fix tests * relock * relock * more client adjustments * pattern prefill * some tts agent fixes * fix ai assist cond * tts nodes * fix config retrieval * assign voice node and fixes * sim suite char gen assign voice * fix voice assign template to consider used voices * get rid of auto break repetition which wasn't working right for a while anyhow * linting * generate tts node as string node * linting * voice change on character event * tweak chatterbox max length * koboldai default template * linting * fix saving of existing voice * relock * adjust params of eva default voice * f5tts support * f5tts samples * f5tts support * f5tts tweaks * chunk size per tts api and reorg defaul f5tts voices * chatterbox default voice reog to match f5-tts default voices * voice library ux polish pass * cleanup * f5-tts tweaks * missing samples * get rid of old save cmd * add chatterbox and f5tts * housekeeping * fix some issues with world entry editing * remove cruft * replace exclamation marks * fix save immutable check * fix replace_exclamation_marks * better error handling in websocket plugins and fix issue with saves * agent config save on dialog close * ctrl click to disable / enable agents * fix quick config * allow modifying response size of focal requests * sim suite set goal always sets story intent, encourage calling of set goal during simulation start * allow setting of model * voice param tweaks * tts tweaks * fix character card load * fix note_on_value * add mixed speaker_separation mode * indicate which message the audio is for and provide way to stop audio from the message * fix issue with some tts generation failing * linting * fix speaker separate modes * bad idea * linting * refactor speaker separation prompt * add kimi think pattern * fix issue with unwanted cover image replacemenT * no scene analysis for visual promp generation (for now) * linting * tts for context investigation messages * prompt tweaks * tweak intro * fix intro text tts not auto playing sometimes * consider narrator voice when assigning voice tro a character * allow director log messages to go only into the director console * linting * startup performance fixes * init time * linting * only show audio control for messagews taht can have it * always create story intent and dont override existing saves during character card load * fix history check in dynamic story line node add HasHistory node * linting * fix intro message not having speaker separation * voice library character manager * sequantial and cancelable auto assign all * linting * fix generation cancel handling * tooltips * fix auto assign voice from scene voices * polish * kokoro does not like lazy import * update info text * complete scene export / import * linting * wording * remove cruft * fix story intent generation during character card import * fix generation cancelled emit status inf loop * prompt tweak * reasoning quick toggle, reasoning token slider, tooltips * improved reasoning pattern handling * fix indirect coercion response parsing * fix streaming issue * response length instructions * more robust streaming * adjust default * adjust formatting * litning * remove debug output * director console log function calls * install cuda script updated * linting * add another step * adjust default * update dialogue examples * fix voice selection issues * what's happening here * third time's the charm? * Vite migration (#207) * add vite config * replace babel, webpack, vue-cli deps with vite, switch to esm modules, separate eslint config * change process.env to import.meta.env * update index.html for vite and move to root * update docs for vite * remove vue cli config * update example env with vite * bump frontend deps after rebase to 32.0 --------- Co-authored-by: pax-co <Pax_801@proton.me> * properly referencer data type * what's new * better indication of dialogue example supporting multiple lines, improve dialogue example display * fix potential issue with cached scene anlysis being reused when it shouldn't * fix character creation issues with player character toggle * fix issue where editing a message would sometimes lose parts of the message * fix slider ux thumb labels (vuetify update) * relock * narrative conversation format * remove planning step * linting * tweaks * don't overthink * update dialogue examples and intro * dont dictate response length instructions when data structures are expected * prompt tweaks * prompt tweaks * linting * fix edit message not handling : well * prompt tweaks * fix tests * fix manual revision when character message was generated in new narrative mode * fix issue with message editing * Docker packages relese (#204) * add CI workflow for Docker image build and MkDocs deployment * rename CI workflow from 'ci' to 'package' * refactor CI workflow: consolidate container build and documentation deployment into a single file * fix: correct indentation for permissions in CI workflow * fix: correct indentation for steps in deploy-docs job in CI workflow * build both cpu and cuda image * docs * docs * expose writing style during state reinforcement * prompt tweaks * test container build * test container image * update docker compose * docs * test-container-build * test container build * test container build * update docker build workflows * fix guidance prompt prefix not being dropped * mount tts dir * add gpt-5 * remove debug output * docs * openai auto toggle reasoning based on model selection * linting --------- Co-authored-by: pax-co <123330830+pax-co@users.noreply.github.com> Co-authored-by: pax-co <Pax_801@proton.me> Co-authored-by: Luis Alexandre Deschamps Brandão <brandao_luis@yahoo.com> |
||
|
veguAI
|
9eb4c48d79 |
0.31.0 (#193)
* some prompt cleanup * prompt tweaks * prompt tweaks * prompt tweaks * set 0.31.0 * relock * rag queries add brief analysis * brief analysis before building rag questions * rag improvements * prompt tweaks * address circular import issues * set 0.30.1 * docs * numpy to 2 * docs * prompt tweaks * prompt tweak * some template cleanup * prompt viewer increase height * fix codemirror highlighting not working * adjust details height * update default * change to log debug * allow response to scale to max height * template cleanup * prompt tweaks * first progress for installing modules to scene * package install logic * package install polish * package install polish * package install polish and fixes * refactor initial world state update and expose setting for it * add `num` property to ModuleProperty to control order of widgets * dynamic storyline package info * fix issue where deactivating player character would cause inconsistencies in the creative tools menui * cruft * add openrouter support * ollama support * refactor how model choices are loaded, so that can be done per client instance as opposed to just per client type * set num_ctx * remove debug messages * ollama tweaks * toggle for whether or not default character gets added to blank talemate scenes * narrator prompt tweaks and template cleanup * cleanup * prompt tweaks and template cleanup * prompt tweaks * fix instructor embeddings * add additional error handling to prevent broken world state templates from breaking the world editor side menu * fix openrouter breaking startup if not configured * remove debug message * promp tweaks * fix example dialogue generation no longer working * prompt tweaks and better showing of dialogue examples in conversation instructions * prompt tweak * add initial startup message * prompt tweaks * fix asset error * move complex acting instructions into the task block * fix content input socket on DynamicInstructions node * log.error with traceback over log.exception since that has a tendency to explode and hang everything * fix usse with number nodes where they would try to execute even if the incoming wire wasnt active * fix issue with editor revision events missing template_vars * DynamicInstruction node should only run if both header and content can resolve * removed remaining references to 90s adventure game writing style * prompt tweaks * support embeddings via client apis (koboldcpp) * fix label on client-api embeddings * fix issue where adding / removing an embedding preset would not be reflected immediately in the memory agent config * remove debug output * prompt tweaks * prompt tweaks * autocomplete passes message object * validate group names to be filename valid * embedded winsows env installs and up to poetry2 * version config * get-pip * relock * pin runpod * no longer needed * remove rapidfuzz dependency * nodejs directly into embedded_node without a versioned middleman dir - also remove defunct local-tts install script * fix update script * update script error handling * update.bat error handling * adjust wording * support loading jinja2 templates node modules in templates/modules * update google model list * client t/s and business indicator - also switch all clients to async streaming * formatting * support coercion for anthropic / google switch to the new google genai sdk upgrade websockets * more coercion fixes * gracefully handle keyboard interrupt * EmitSystemMessage node * allow visual prompt generation without image generation * relock * chromadb to v1 * fix error handling * fix issue where adding client model list would be empty * supress pip install warnings * allow overriding of base urls * remove key from log * add fade effect * tweak request info ux * further clarification of endpoint override api key * world state manager: fix issue that caused max changes setting to disappear from character progress config * fix issue with google safety settings off causing generation failures * update to base url should always reset the client * getattr * support v3 chara card version and attempt to future proof * client based embeddings improvements * more fixes for client based embeddings * use client icon * history management tools progress * history memory ids fixed and added validation * regenerate summary fixes * more history regeneration fixes * fix layered history gen and prompt twweaks * allow regeneration of individual layered history entries * return list of LayeredArchiveEntry * reorg for less code dupelication * new scene message renderer based on marked * add inspect functionality to history viewer * message if no history entries yet * allow adding of history entries manually * allow deletion of history * summarization unslop improvements * fix make charcter real action from worldstate listing * allow overriding length in all context generation isntructioon dialogs * fix issue where extract_list could fail with an unhandled error if the llm response didnt contain a list * update whats'new * fix issues with the new history management tools * fix check * Switch dependency handling to UV (#202) * Migrate from Poetry to uv package manager (#200) * migrate from poetry to uv package manager * Update all installation and startup scripts for uv migration * Fix pyproject.toml for uv - allow direct references for hatchling * Fix PR feedback: Restore removed functionality - Restored embedded Python/Node.js functionality in install.bat and update.bat - Restored environment variable exposure in docker-compose.yml (CUDA_AVAILABLE, port configs) - Fixed GitHub Actions branches (main, prep-* instead of main, dev) - Restored fail-fast: false and cache configuration in test.yml These changes preserve all the functionality that should not be removed during the migration from Poetry to uv. --------- Co-authored-by: Ztripez von Matérn <ztripez@bobby.se> * remove uv.lock from .gitignore * add lock file * fix install issues * warn if unable to remove legacy poetry virt env dir * uv needs to be explicitly installed into the .venv so its available * third time's the charm? * fix windows install scripts * add .venv guard to update.bat * call :die * fix docker venv install * node 21 * fix cuda install * start.bat calls install if needed * sync start-local to other startup scripts * no need to activate venv --------- Co-authored-by: Ztripez <reg@otherland.nu> Co-authored-by: Ztripez von Matérn <ztripez@bobby.se> * ignore hfhub symlink warnings * add openrouter and ollama mentions * update windows install documentation * docs * docs * fix issue with memory agent fingerprint * removing a client that supports embeddings will also remove any embedding functions it created * on invalid embeddings reset to default * docs * typo * formatting * docs * docs * install package * adjust topic * add more obvious way to exit creative mode * when importing character cards immediately persist a usable save after the restoration save --------- Co-authored-by: Ztripez <reg@otherland.nu> Co-authored-by: Ztripez von Matérn <ztripez@bobby.se> |
||
|
veguAI
|
113553c306 |
0.29.0 (#167)
* set 0.29.0 * tweaks for dig layered history (wip) * move director agent to directory * relock * remove "none" from dig_layered_history response * determine character development * update character sheet from character development (wip) * org imports * alert outdated template overrides during startup * editor controls normalization of exposition * dialogue formatting refactor * fix narrator.clean_result forcing * regardless of editor fix exposition setting * move more of the dialogue cleanup logic into the editor fix exposition handlers * remove cruft * change ot normal selects and add some margin * move formatting option up * always strip partial sentences * separates exposition fixes from other dialogue cleanup operations, since we still want those * add novel formatting style * honor formatting config when no markers are supplied * fix issue where sometimes character message formatting would miss character name * director can now guide actors through scene analysis * style fixes * typo * select correct system message on direction type * prompt tweaks * disable by default * add support for dynamic instruction injection and include missing guide for internal note usage * change favicon and also indicate business through favicon * img * support xtc, dry and smoothing in text gen webui * prompt tweaks * support xtc, dry, smoothing in koboldcpp client * reorder * dry, xtc and smoothing factor exposed to tabby api client * urls to third party API documentation * remove bos token * add missing preset * focal * focal progress * focal progress and generated suggestions progress * fix issue with discard all suggestions * apply suggestions * move suggestion ux into the world state manager * support generation options for suggestion generation * unused import * refactor focal to json based approach * focal and character suggestion tweaks * rmeove cruft * remove cruft * relock * prompt tweaks * layout spacing updates * ux elements for removal of scenes from quick load menu * context investigation refactor WIP * context investigation refactor * context investigation refactor * context investigation refactor * cleanup * move scene analysis to summarizer agent * remove deprecated context investigation logic * context investigation refactor continued - split into separate file for easier maint * allow direct specification of response context length * context investigation and scene analyzation progress * change analysis length config to number * remove old dig-layered-history templates * summarizer - deep analysis is only available if there is layered history * move world_state agent to dedicated directory * remove unused imports * automatic character progression WIP * character suggestions progress * app busy flag based on agent business * indicate suggestions in world state overview * fix issue with user input cleanup * move conversation agent to a dedicated submodule * Response in action analyze_text_and_extract_context is too short #162 * move narrator agent to its own submodule * narrator improvements WIP * narration improvements WIP * fix issue with regen of character exit narration * narration improvements WIP * prompt tweaks * last_message_of_type can set max iterations * fix multiline parsing * prompt tweaks * director guide actors based of scene analysis * director guidance for actors * prompt tweaks * prompt tweaks * prompt tweaks * fix automatic character proposals not propagating to the ux * fix analysis length * support director guidance in legacy chat format * typo * prompt tweaks * prompt tweaks * error handling * length config * prompt tweaks * typo * remove cruft * prompt tweak * prompt tweak * time passage style changes * remove cruft * deep analysis context investigations honor call limit * refactor conversation agent long term memory to use new memory rag mixin - also streamline prompts * tweaks to RAG mixin agent config * fix narration highlighting * context investgiation fixes director narration guidance summarization tweaks * direactor guide narration progress context investigation fixes that would cause looping of investigations and failure to dig into the correct layers * prompt tweaks * summarization improvements * separate deep analysis chapter selection from analysis into its own prompt * character entry and exit * cache analysis per subtype and some narrator prompt tweaks * separate layered history logic into its own summarizer mixin and expose some additional options * scene can now set an overral writing style using writing style templates narrator option to enable writing style * narrate query writing style support * scene tools - narrator actions refactor to handler and own component * narrator query / look at narrations emitted as context investigation messages refactor context investigation messaage display scene message meta data object * include narrative direction * improve context investigation message prompt insert * reorg supported parameters * fix bug when no message history exists * WIP make regenerate work nicely with director guidance * WIP make regenerate work nicely with director guidance * regenerate conversation fixes * help text * ux tweaks * relock * turn off deep analysis and context investigations by default * long term memory options for director and summarizer * long term memory caching * fix summarization cache toggle not showing up in ux * ux tweaks * layered history summarization includes character information for mentioned characters * deepseek client added * Add fork button to narrator message * analyze and guidance support for time passage narration * cache based on message fingerprint instead of id * configurable system prompts WIP * configurable system prompts WIP * client overrides for system prompts wired to ux * system prompt overhaul * fix issue with unknown system prompt kind * add button to manually request dynamic choices from the director move the generate choices logic of the director agent to its own submodule * remove cruft * 30 may be too long and is causing the client to disappear temporarly * suppoert dynamic choice generate for non player characters * enable `actor` tab for player characters * creator agent now has access to rag tools improve acting instruction generation * client timeout fixes * fix issue where scene removal menu stayed open after remove * expose scene restore functionality to ux * create initial restore point * fix creator extra-context template * didn't mean to remove this * intro scene should be edited through world editor * fix alert * fix partial quotes regardless of editor setting director guidance for conversation reminds to put speech in quotes * fix @ instructions not being passed through to director guidance prompt * anthropic mode list updated * default off * cohere model list updated * reset actAs on next scene load * prompt tweaks * prompt tweaks * prompt tweaks * prompt tweaks * prompt tweaks * remove debug cruft * relock * docs on changing host / port * fix issue with narrator / director actiosn not available on fresh install * fix issue with long content classification determination result * take this reminder to put speech into quotes out for now, it seems to do more harm than good * fix some remaining issues with auto expositon fixes * prompt tweaks * prompt tweaks * fix issue during reload * expensive and warning ux passthrough for agent config * layered sumamry analysation defaults to on * what's new info block added * docs * what's new updated * remove old images * old img cleanup script * prompt tweaks * improve auto prompt template detection via huggingface * add gpt-4o-realtime-preview add gpt-4o-mini-realtime-preview * add o1 and o3-mini * fix o1 and o3 * fix o1 and o3 * more o1 / o3 fixes * o3 fixes |