* feat: add bulk extension field updates with UNSET_VALUE sentinel for key deletion
- Add `UNSET_VALUE` sentinel constant to signal complete field removal from character cards
- Add `writeExtensionFieldBulk()` function to update extension fields across multiple characters in a single API call
- Add `deleteValueByPath()` utility function to remove nested object keys by dot-path
- Update `writeExtensionField()` to support `UNSET_VALUE` for deleting extension keys
- Extend `/api/characters/merge-attributes
* Revert package-lock.json changes
* Allow null values in merge-attributes filter path validation
Change filter.path existence check to only skip on undefined, not null. This allows merging attributes when the existing value is explicitly null, treating null as a valid value rather than absence of a value.
* fix: share forbiddenRegExp between modules
* feat: add writeExtensionFieldBulk and UNSET_VALUE constant to getContext
* Update src/endpoints/characters.js
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix: validate for .png extension
* Update public/scripts/extensions.js
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor: extract shouldSkip logic as a function param to avoid double parsing
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
In group chats, only include reasoning from the currently generating character instead of all group members. This prevents reasoning from other characters being injected into the prompt context when generating responses.
- Filter reasoning in coreChat loop based on message author matching name2
- Filter reasoning in setOpenAIMessages based on message author matching name2
- Add isOtherGroupMember check before adding reasoning to messages
Raises the maximum allowed Top K sampling parameter from 200 to 500 in both the range slider and number input controls to support higher token selection limits.
* Fix internationalization issues
* fix(i18n): update world creation key and zh-cn translations
Changes the i18n key for the world editor's "New" button to "Create" to
avoid conflict with listing filters. Updates the Chinese translation
for the "New" key to "最新" (Latest).
* fix(i18n): correct translation key for world creation button
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* Add range and silent parameters to /persona-sync command with optional confirmation suppression
- Add optional start/end range parameters to syncUserNameToPersona() function
- Add silent parameter to suppress confirmation popup when needed
- Update /persona-sync slash command to accept range argument (index or range string) and named silent argument
- Parse range using stringToRange() utility, default to full chat if not provided
- Update confirmation message to reflect whether syncing all messages or specified
* Add `from` named argument to /persona-sync command for filtering by persona name
- Add `from` named argument to filter messages by persona name (case-insensitive)
- Rename `silent` argument to `quiet` with inverted default (true) for consistency
- Add userMessageNamesEnumProvider() to provide autocomplete for existing user message names in chat
- Update syncUserNameToPersona() to accept nameFilter parameter and filter messages accordingly
- Update confirmation message to reflect name filtering when
* Add async/await wrapper to sync_name_button click handler for proper promise handling
Function now has arguments, so using just the function as the event is shown as wrong usage
* Post-merge imports fix
* Use canonical command name in examples
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* enable interleaved tool reasoning for custom OpenAI-compat endpoints
Add chat_completion_sources.CUSTOM to interleaved_reasoning_providers so
that local OpenAI-compatible endpoints (e.g. KoboldCPP in Chat Completions
mode) can forward reasoning context in tool-call chains when the user has
configured Interleaved Thinking.
Also expose the Interleaved Thinking UI control for the Custom source so
users can actually opt in — previously the dropdown was hidden behind a
data-source="openrouter" guard.
The custom streaming path already correctly accumulates delta.reasoning_content
from streaming chunks; this change only removes the provider gate that was
silently discarding that data before it reached the API payload.
* don't override invocation reasoning with prior-turn assistant reasoning
When an invocation already has its own reasoning captured at execution
time, preserve it instead of replacing it with previousAssistantReasoning
from the backward scan. The override was correct when invocations never
carried their own reasoning, but now that the custom/openrouter paths
capture per-invocation reasoning, the unconditional replacement caused
all tool calls in a chain to receive the same stale reasoning from an
earlier unrelated assistant turn.
Fall back to previousAssistantReasoning only when clone.reasoning is empty.
* Add StreamingDisplay class for live LLM generation output with floating toast panel
- Add StreamingDisplay class to show streaming reasoning and content in a floating toast panel
- Extract createModelIcon() helper from insertSVGIcon() for reusable API/model icon creation
- StreamingDisplay automatically appends inside topmost open dialog (same pattern as fixToastrForDialogs)
- Add CSS with fade-in animation, pulsating activity indicator, and separate reasoning/content sections
- Support optional model icon in header
* Add ConnectionManagerRequestService.getProfileIcon() method for retrieving profile API icons
- Add static getProfileIcon() method to ConnectionManagerRequestService
- Returns HTMLImageElement created via createModelIcon() for a given profile's API/model
- Accepts optional profileId parameter, defaults to currently selected profile
- Returns null if Connection Manager is disabled, profile not found, or profile has no API
- Import createModelIcon from script.js
* Use animation_duration directly in hide() and CSS transition instead of constant
- Remove ANIMATION_DURATION_MS constant and use animation_duration directly in hide() method
- Replace hardcoded 0.3s CSS transitions with CSS variable var(--animation-duration, 125ms)
- Read animation_duration value inline in hide() for accurate timing
* Add /genstream slash command with live streaming display and reasoning support
- Add /genstream slash command that generates text via Connection Manager with live streaming UI
- Add formatReasoning() helper function (inverse of parseReasoningFromString) to format reasoning/content into template-wrapped strings
- Add connectionProfiles enum provider for profile selection in slash commands
- StreamingDisplay: add delay parameter to hide() method (default 1000ms) to show final result before dismiss
* Add /reasoning-format slash command to format reasoning and content into template-wrapped strings
- Add /reasoning-format (alias: /format-reasoning) slash command that wraps reasoning/content using Reasoning Formatting settings
- Accept required 'reasoning' named argument and optional unnamed 'content' argument
- Validate that prefix/suffix are configured before formatting
- Return formatted string via formatReasoning() helper for use with /reasoning-parse
- Show warning toasts if prefix/suffix missing
* Rename /genstream command to /profile-genstream and move to appropriate module
* Apply messageFormatting to StreamingDisplay reasoning and content text for proper rendering
- Import messageFormatting from script.js
- Replace textContent with innerHTML using messageFormatting() in updateReasoning() and updateText()
- Pass isSystem=true for reasoning, isSystem=false for content to match formatting expectations
- Update css to utilize pre-formatted paragraphs correctly
* Strip auto-added quotes from <q> tags in StreamingDisplay and add 'mes_text' class for consistent chat message formatting
- Add CSS rules to remove browser-default quotes from <q> tags in reasoning and content sections
- Add 'mes_text' class to textContent div to match chat message formatting behavior
- Prevents double quotes when messageFormatting already adds them via <q> tags
* Add minimize/close buttons and complete state to StreamingDisplay with configurable auto-hide
- Add minimize button to collapse/restore content sections while keeping header visible
- Add close button to manually dismiss display (generation continues in background)
- Replace CSS pseudo-element with explicit LED indicator element for better state control
- Add complete() method to mark generation done: changes LED from pulsing orange to solid green
- Add configurable auto-hide delay after completion
* Add stop button to StreamingDisplay with abort support and onStop/onComplete closures for /profile-genstream
- Add stop button to StreamingDisplay when onStop handler is provided
- Add markStopped() method with solid red LED state indicator
- Add AbortController integration to /profile-genstream for request cancellation
- Add onStop and onComplete closure arguments to /profile-genstream command
- Update complete() method signature to use options object with label and delay
- Disable stop button immediately
* Position StreamingDisplay above bottom form block using CSS variable with fallback
- Change bottom positioning from fixed 20px to dynamic calculation
- Use max() to position above --bottomFormBlockSize + 5px or minimum 20px
- Ensures StreamingDisplay doesn't overlap with bottom UI elements
* Rename /profile-genstream arguments for clarity: label→generating, completedLabel→completed, hideDelay→delay
- Rename `label` argument to `generating` to better reflect its purpose as the in-progress state label
- Rename `completedLabel` to `completed` for consistency and brevity
- Rename `hideDelay` to `delay` for simpler naming
- Update all internal references and variable names to match new argument names
- Update argument descriptions and default values accordingly
* Remove variable resolution from /profile-genstream arguments: system, length, and delay
- Remove ARGUMENT_TYPE.VARIABLE_NAME from typeList for system, length, and delay arguments
- Replace resolveVariable() calls with direct argument access for system, length, and delay
- Simplify type checking to use typeof directly on args properties
- Maintain existing default values and validation logic
* Add warning toast and early return when connection profile not found in /profile-genstream
- Display toastr warning when fuzzy search fails to find matching profile
- Return empty string to prevent execution with invalid profile
- Improves user feedback for incorrect profile names or IDs
* Extract buildResultText() helper in /profile-genstream to return partial results when stopped
- Add buildResultText() helper function to centralize result formatting logic
- Return partial generated text when user stops generation instead of empty string
- Reuse buildResultText() for both stopped and completed states
- Maintains consistent reasoning formatting in both cases
* fix lint
* Update documentation to reflect argument name change from hideDelay to delay
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* feat: emit PERSONA_CREATED event when duplicating a persona
Adds PERSONA_CREATED event emission in duplicatePersona() to notify listeners when a persona is duplicated. Includes the new avatarId, name, and duplicatedFromAvatarId in the event payload.
* fix: event data and execution order
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* Add 'clean' extension hook support with optional cleanup on delete
- Add hasExtensionHook() helper to check if an extension defines a specific manifest hook
- Add 'clean' hook type to callExtensionHook() JSDoc
- Add clean button to extension UI when 'clean' hook is present
- Add onCleanClick() handler to run clean hook with confirmation
- Add cleanExtension() function to execute clean hook and reload page
- Modify deleteExtension() to optionally run clean hook before deletion
- Show cleanup checkbox on extension delete popup
* fix lint
Remove unused `callGenericPopup` import from extensions.js
* Force save settings before page reload in extension clean and delete operations
Add explicit saveSettings() calls in cleanExtension() and deleteExtension() to prevent race conditions where clean/delete hooks might update settings that get lost during the subsequent page reload.
* Remove admin permission check from extension clean operation
The clean hook is extension-defined and may not require admin privileges. Permission checks should be handled by the extension's clean hook implementation if needed, rather than enforcing a blanket restriction at the UI level.
* fix: show clean button for built-ins
* Update hasExtensionHook to support built-in extensions
* Revert "Update hasExtensionHook to support built-in extensions"
This reverts commit 31be55ea66430ffe6a8d149d519b3d6d149da9ea.
* Revert "fix: show clean button for built-ins"
This reverts commit 5f86fec70c2b7d5cd99e4dee7f14af5a3372d58f.
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* allow grammar to keep place during continue
* lint
* fix: require kai flag for grammar to be set
* fix: don't retain if grammar is empty
* fix: default to undefined if not retain
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
Adds optional `disabled` boolean property to PopupInputConfiguration JSDoc and implements it across all input types (checkbox, text, textarea, number). When set to true, the input element will be rendered in a disabled state.
Returns a structuredClone of the loaded manifest for the given extension name.
Accepts either the short name or full internal key (third-party/...).
Returns null if the extension is not registered.
* feat: add /regex-state command to check script enabled status
Adds a new slash command that returns 'on' or 'off' to indicate whether a regex script is currently enabled or disabled.
* refactor: change /regex-state return values from 'on'/'off' to 'true'/'false'
Updates the /regex-state slash command to return boolean string values ('true'/'false') instead of ('on'/'off') for better consistency with standard boolean conventions.
* fix: include model field in sd.cpp SDAPI requests and preserve URL path
The sd.cpp integration overwrites the URL pathname when constructing
requests, which breaks proxy servers like llama-swap that use path-based
routing (e.g. /upstream/model-name). Additionally, the model field was
never included in SDAPI requests, which is required by llama-swap to
route requests to the correct backend.
Changes:
- Server: Append to URL pathname instead of overwriting (same pattern as #5178)
- Server: Pass model field through to sd-server payload
- Client: Add model name text input for sd.cpp source settings
- Client: Send model name in generate request payload
* fix: fetch models from server and populate standard Model dropdown
Instead of a separate text input for the model name, fetch the model
list from the sd.cpp server's /v1/models endpoint and populate the
standard Model dropdown. This provides a seamless experience where
users just pick a model from the dropdown like any other source.
Works with both standalone sd-server and proxy servers like llama-swap
that expose multiple models via the OpenAI-compatible models endpoint.
* fix: don't send clip_skip=1 to sd.cpp, it produces blank images
sd-server generates blank white images when clip_skip is set to 1.
Since clip_skip=1 means 'use all CLIP layers' (the default behavior),
only send the parameter when it's > 1.
* Fix eslint
* Replace string appends with urlJoin
* fix: convert URL strings to URL objects in sdcpp routes
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
Extends the #logDeprecated helper to accept and log the arguments passed to deprecated methods, providing better debugging context for migration. Updates all deprecated method calls (get, has, registerMacro, unregisterMacro) to forward their arguments to the deprecation logger.
Prevents invalid CSS by skipping rules containing `::` (pseudo-elements like `::before`, `::after`, `::-webkit-scrollbar`) when converting `:hover` to `:focus-visible`, as pseudo-elements cannot receive focus.
Previously only exact 'en' was accepted as fallback when a language wasn't found. Now accepts any English variant (en-US, en-AU, etc.) to prevent spurious warnings for valid English locales.
* feat: add Workers AI text embeddings and multimodal captioning
Extends the Cloudflare Workers AI integration to the vectors and
caption extensions.
Embeddings: adds workers_ai source to the vectors extension using the
OpenAI-compatible /v1/embeddings endpoint, with dynamic model listing
from the Cloudflare model search API.
Captioning: adds workers_ai as a multimodal caption API with dynamic
vision model discovery via the multimodal-models endpoint.
* Add logo svg
* Refactor caption dropdown population
* Fix order of sources
* feat: add error handling for missing Workers AI account ID
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* refactor(tts): remove redundant 4-space codeblock filter
- Deleted the regex that removed lines starting with four spaces.
- Original intent was to strip "indented code blocks" (Markdown legacy syntax).
- In practice, SillyTavern already handles explicit code fences (```...``` and ~~~...~~~).
- Indented code blocks are rarely used and the regex caused unnecessary text loss in normal messages.
- Simplifies codeblock skipping logic and avoids accidental removal of valid content.
* Remove commented out code
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* feat: add entry preview to world info deletion confirmation dialog
Displays entry comment or first two lines of content in the deletion confirmation popup to help users verify they're deleting the correct entry.
* fix: sanitize world info entry preview text in deletion confirmation dialog
Adds DOMPurify sanitization to the entry preview text displayed in the deletion confirmation popup to prevent potential XSS vulnerabilities from unsanitized user content.
* DOMPurify -> escapeHtml
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* feat: add Cloudflare Workers AI provider
Adds support for Cloudflare Workers AI using its OpenAI-compatible API.
Workers AI-specific stuff includes:
- Model list fetching and capabilities detection
- Tokenizer auto-detection for typical hosted model families
- Streaming not supported when using structured output
Closes#5305
* Make the entire header clickable
* Add missing samplers
* Fix non-streaming reasoning parsing
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* Skip TTS for voices explicitly set to disabled (fixes#4970)
* Always show disabled message in commands and fix restoring voice map UI
* Always show a message on manual TTS trigger
* Fix null current job on disabled
* Adjust type annotation
* Force update worker when disabled play is attempted
* Treat audio control queue as manual
* Update TTS message processing to include manual flag
* Don't show toast if was already shown by manual playback
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* Use custom init script instead of postinstall
* Revert changes to start scripts in src\electron
* Add global data to content manager
* Add migration for public overrides and user.css location update
* Update npm publish workflow to use 'omit=dev' flag in npm ci commands
* Rename user.css readme file
* Fix indentation in userCssMiddleware function
* Add directory creation for content target
* Restore template compile location
* Move stylesheet up in index.json
* Use path.resolve for user.css file path in userCssMiddleware
* Correct capitalization in "Not Found" error page title and heading
* Remove init run from startup scripts
* Simplify user CSS file path resolution
* Update userCssMiddleware comment
* fix: require long press to open swipe picker on phones
* fix: clarify parameter description in assignLorebookToChat function
* fix: update event parameter type in onSwipeCounterClick to include TouchEvent
* fix: update event parameter types in onSwipeCounterClick and addLongPressEvent
* fix: return Error objects from invokeFunctionTool and create error invocations
invokeFunctionTool previously called .toString() on caught errors,
converting them to plain strings. This made the instanceof Error
check in invokeFunctionTools dead code.
Changes:
- Return Error objects directly from invokeFunctionTool
- Create error invocations with error: true flag when tools fail
- Record failed stealth tools in stealthCalls
- Preserve signature/reasoning on error invocations
- Add error field to ToolInvocation typedef
* fix: use Error.toString to avoid behavioral changes
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* feat(secrets): update readSecret function to accept optional secret ID
* add secret_id to ConnectionManagerRequestService payload
* fix: pass secret_id for Text Completion types
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* fix: conditionally include secrets in user data backup
* feat: add full data backup toggle
* 418 -> 403
I'm not a teapot
* Distinguish fails from disabled
* feat(ui): add popup to jump to a specific swipe
Adds a new "Jump to swipe history" button to message actions and makes
the swipe counter clickable on the latest message. This opens a
searchable popup allowing users to quickly find and jump to a
specific alternate swipe without having to click through them
sequentially.
- Adds a searchable swipe picker popup menu
- Makes the swipe counter interactive when multiple swipes exist
- Adds a dedicated swipe picker button to message controls
* fix(ui): hide swipe picker when there are no swipes
Ensures the newly added swipe picker button is only shown when a
message has multiple swipes available. It explicitly hides the
button for non-swipeable messages or messages with a single swipe.
* feat(ui): redesign swipe picker with direct id input
Replace text-based search with a numeric input for direct swipe navigation.
Update popup layout with a sticky header and improved scrolling behavior.
Sync input value with the currently selected swipe in the list.
Refactor styling to align with chat selection components.
* feat(ui): allow branching from specific swipes via picker
- Enable swipe picker for historical messages to inspect alternate swipes
- Add branch button to picker entries to create new chats from specific swipes
- Update saveChat and createBranch to accept chat snapshots
- Restrict swipe jumping to the active message only
* refactor(logic): consolidate swipe sync logic and simplify helpers
Update `syncSwipeToMes` to accept a target message object, enabling its
use in the bookmarks module and removing the duplicate
`applySwipeToSnapshot` function.
Also simplify `canOpenSwipePickerForMessage` and
`canJumpToSwipeForMessage` signatures by removing the redundant message
parameter.
* refactor(a11y): support dynamic roles via classes
Introduce a managed role system in the accessibility script to handle
elements that dynamically gain or lose interactive states. The mutation
observer now watches for class attribute changes and automatically
applies or clears roles (e.g., `role="button"`) using active selectors.
Updated the swipe counter to rely on this centralized system by toggling
an `.interactable` class instead of manually modifying tabindex and role
attributes. Removed the redundant 'Enter' keydown handler for the swipe
counter to prevent duplicate trigger events.
* fix(ui): compute missing token counts in swipe picker
Update renderSwipeList to asynchronously calculate token counts when
missing from swipe metadata. Introduce SWIPE_SOURCE.SWIPE_PICKER to
correctly identify swipes triggered from the picker and bypass
generation checks.
* feat(ui): enable deleting specific swipes via swipe picker
- Adds a delete button to swipe picker entries, allowing removal of specific message versions.
- Refactors deletion logic to handle removing non-current swipes without triggering animations and correctly updates indices.
- Includes confirmation dialogs and improves input focus behavior.
* refactor:Delete process inline to button click processor
* feat: universal swipe inspection and picker improvements
- Permit opening the swipe browser on any chat entry to review past generations.
- Parallelize the retrieval of token statistics to speed up list rendering.
- Format message metrics (length and tokens) into a single, concise string.
- Update the `getBranchChatSnapshot` API to accept an options object.
- Register swipe list items as interactable elements for keyboard control.
- Apply styling to prevent text highlighting on picker entries.
* fix:remove unused CSS
* fix: fix disabled styling for swipe delete button
Remove tooltips and prevent hover animations or glow effects when the
delete button is disabled in the swipe picker. Update CSS to enforce
default cursor and fixed opacity on hover for the disabled state.
* remove: Unused CSS
* Extract swipe-picker.js module
* Revert to manual ARIA role management
* Avoid scrollIntoView and scroll on open
* Fix keyboard interaction in past chats menu
* Fix a11y attribute
* fix: call refreshSwipeButtons when deleting not selected swipe
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* feat: add finalizeIntermediaryMessage for streaming tool call chains
Add a lighter finalization method on StreamingProcessor that runs
before tool invocation. This ensures CHARACTER_MESSAGE_RENDERED is
emitted for streamed text messages before the tool call chain
continues, allowing extensions (TTS, image forwarding, etc.) to
process intermediary messages.
finalizeIntermediaryMessage performs essential processing:
- Code block styling (addCopyToCodeBlocks)
- Reasoning handler finalization
- Logprobs saving
- Image attachment processing
- Reasoning signature storage
- MESSAGE_RECEIVED and CHARACTER_MESSAGE_RENDERED events
Without the heavier onFinishStreaming operations:
- UI unlock (markUIGenStopped)
- Auto-swipe
- Sound playback
- Chat saving
- Swipe counter update
* fix: return Error objects from invokeFunctionTool and create error invocations
invokeFunctionTool previously called .toString() on caught errors,
converting them to plain strings. This made the 'instanceof Error'
check in invokeFunctionTools dead code - errors silently became
successful invocations, showToolCallError never fired, and the
result.errors array was always empty.
Changes:
- Return Error objects directly from invokeFunctionTool
- Create error invocations with error: true flag when tools fail
- Add error field to ToolInvocation typedef
* fix: record failed stealth tools in stealthCalls and preserve signature/reasoning on error invocations
When a stealth tool errors, its name was not added to stealthCalls, causing
shouldStopGeneration to evaluate as false. This led to an incorrect recursive
Generate('normal') call instead of stopping generation as the stealth tool
contract requires ('no follow-up generation').
Also preserve toolCall.signature and reasoningText on error invocations to
match the success path, preventing Gemini/OpenRouter multi-turn tool context
from breaking when a tool call fails.
* Clarify method comments
* fix: initialize error property in ToolInvocation to false
* Apply review suggestions
* Make options object required
* fix: remove unnecessary return statement in updateSwipeCounter method
* refactor: split tool call error handling into separate PR (#5351)
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
* renaming a lorebook prompts to update existing links
* used suggested api and logic
* add world property to shallow function
* Fix type error in assignLorebookToChat invoke
* Remove debug console logs
* Fix activeCharacterUpdated
* Extract updateWorldInfoLinks into a func
* Invert if for an early return
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
Cohee