\n {mode === \"create\" ? \"New Chat\" : \"Rename Chat\"}\n

\n\n

\n \n Chat Name\n \n {\n setName(e.target.value);\n setValidationError(\"\");\n }}\n placeholder=\"e.g., Work Ideas, Journal, Research...\"\n disabled={isLoading}\n className=\"w-full px-3 sm:px-4 py-3 min-h-[44px] rounded-md bg-vibe-300 border border-vibe-250 text-white placeholder-vibe-200 focus:outline-none focus:ring-1 focus:ring-primary-300 disabled:opacity-50 transition-colors\"\n autoFocus\n />\n {displayError && (\n

{displayError}

\n )}\n

\n\n

\n \n Cancel\n \n \n {isLoading ? \"Saving...\" : \"Save\"}\n \n

\n );\n}\n",[],[194,201],{"_106":210,"_97":211,"_165":212,"_167":195,"_168":150,"_169":213,"_171":121,"_108":214,"_102":215,"_173":205,"_175":216,"_176":22,"_177":-5,"_178":217,"_100":218},"93580f72-de08-11f0-9022-42dde27851f2","EmptyState.tsx","ca11af7c-d7b0-11f0-ae93-42dde27851f2",795,["D",1766278886143],6,"/** @jsxImportSource https://esm.sh/react */\nimport React from \"https://esm.sh/react\";\n\ntype Props = {\n isReady: boolean;\n};\n\nexport function EmptyState({ isReady }: Props) {\n return (\n

{isReady ? \"Send a message to start\" : \"Loading AI...\"}

\n {!isReady && (\n

\n First load downloads takes a bit longer\n

\n )}\n

\n );\n}\n",[],[194,211],{"_106":220,"_97":221,"_165":222,"_167":195,"_168":150,"_169":223,"_171":22,"_108":224,"_102":225,"_173":205,"_175":226,"_176":22,"_177":-5,"_178":227,"_100":228},"dd34a22c-008b-11f1-ba55-42dde27851f2","MessageBubble.tsx","ca068232-d7b0-11f0-b8cb-42dde27851f2",1152,["D",1770073613670],13,"/** @jsxImportSource https://esm.sh/react */\nimport React from \"https://esm.sh/react\";\nimport type { Message } from \"../types.ts\";\nimport { ThinkingBubble } from \"./ThinkingBubble.tsx\";\n\ntype Props = {\n message: Message;\n isStreaming?: boolean;\n streamingPhase?: \"thinking\" | \"response\";\n};\n\nexport function MessageBubble({ message, isStreaming = false, streamingPhase }: Props) {\n const isUser = message.role === \"user\";\n const isAssistant = message.role === \"assistant\";\n\n return (\n

\n {/* Thinking section for assistant messages */}\n {isAssistant && message.thinking && (\n \n )}\n\n {/* Main message content */}\n \n {message.content}\n {/* Streaming cursor for response phase */}\n {isStreaming && streamingPhase === \"response\" && (\n \n )}\n

\n \n );\n}\n",[],[194,221],{"_106":230,"_97":231,"_165":232,"_167":195,"_168":150,"_169":233,"_171":121,"_108":234,"_102":235,"_173":205,"_175":236,"_176":22,"_177":-5,"_178":237,"_100":238},"ab0efd22-edb9-11f0-b895-42dde27851f2","ModelSelector.tsx","c4f7b52c-dd46-11f0-bf67-42dde27851f2",1052,["D",1768004214303],20,"/** @jsxImportSource https://esm.sh/react */\nimport React, { useEffect, useRef, useState } from \"https://esm.sh/react\";\nimport {\n AVAILABLE_MODELS,\n getModelInfo,\n getModelSizeLabel,\n} from \"../lib/models.ts\";\n\ntype Props = {\n selectedModel: string;\n onModelChange: (modelId: string) => void;\n disabled?: boolean;\n isReady?: boolean;\n};\n\nexport function ModelSelector(\n { selectedModel, onModelChange, disabled, isReady }: Props,\n) {\n const [isOpen, setIsOpen] = useState(false);\n const dropdownRef = useRef(null);\n\n const currentModel = getModelInfo(selectedModel);\n const displayLabel = currentModel ? `${currentModel.name}` : \"Select Model\";\n\n // Close dropdown when clicking outside\n useEffect(() => {\n const handleClickOutside = (event: MouseEvent) => {\n if (\n dropdownRef.current &&\n !dropdownRef.current.contains(event.target as Node)\n ) {\n setIsOpen(false);\n }\n };\n\n if (isOpen) {\n document.addEventListener(\"mousedown\", handleClickOutside);\n }\n\n return () => {\n document.removeEventListener(\"mousedown\", handleClickOutside);\n };\n }, [isOpen]);\n\n const handleModelSelect = (modelId: string) => {\n onModelChange(modelId);\n setIsOpen(false);\n };\n\n return (\n

\n {/* Selector Button */}\n !disabled && setIsOpen(!isOpen)}\n disabled={disabled}\n className=\"flex items-center h-fit gap-2 px-2 sm:px-4 py-1 text-sm text-white bg-vibe-300 rounded-[28px] hover:bg-vibe-250 transition-colors disabled:opacity-50 disabled:cursor-not-allowed\"\n type=\"button\"\n >\n {displayLabel}\n \n \n \n \n\n {/* Dropdown Menu */}\n {isOpen && (\n

\n {AVAILABLE_MODELS.map((model) => (\n handleModelSelect(model.id)}\n className={`w-full px-3 sm:px-4 py-3 text-left min-h-[44px] hover:bg-vibe-300 transition-colors ${\n model.id === selectedModel ? \"bg-vibe-300\" : \"\"\n }`}\n type=\"button\"\n >\n

\n {model.name}\n

\n {model.description}\n

\n \n {getModelSizeLabel(model.size)}\n \n \n {model.downloadSize}\n \n

\n \n ))}\n

\n )}\n

\n );\n}\n",[],[194,231],{"_106":240,"_97":241,"_165":242,"_167":195,"_168":150,"_169":243,"_171":121,"_108":244,"_102":245,"_173":205,"_175":246,"_176":22,"_177":-5,"_178":247,"_100":248},"ebbfc84e-edbc-11f0-aa0e-42dde27851f2","OptionsMenu.tsx","3b4198ee-dc6e-11f0-9833-42dde27851f2",1059,["D",1768005611326],35,"/** @jsxImportSource https://esm.sh/react */\nimport React, { useEffect, useRef, useState } from \"https://esm.sh/react\";\nimport { clearMasterPassword } from \"../lib/crypto.ts\";\n\ntype Props = {\n onLock: () => void;\n onDeleteChat: (chatId: string) => void;\n currentChatId: string | null;\n};\n\nexport function OptionsMenu({ onLock, onDeleteChat, currentChatId }: Props) {\n const [isOpen, setIsOpen] = useState(false);\n const [showDeleteConfirm, setShowDeleteConfirm] = useState(false);\n const [showResetConfirm, setShowResetConfirm] = useState(false);\n const menuRef = useRef(null);\n\n useEffect(() => {\n function handleClickOutside(event: MouseEvent) {\n if (menuRef.current && !menuRef.current.contains(event.target as Node)) {\n setIsOpen(false);\n setShowDeleteConfirm(false);\n setShowResetConfirm(false);\n }\n }\n\n if (isOpen) {\n document.addEventListener(\"mousedown\", handleClickOutside);\n return () =>\n document.removeEventListener(\"mousedown\", handleClickOutside);\n }\n }, [isOpen]);\n\n const handleDeleteChatClick = () => {\n if (!currentChatId) return; // No chat selected, do nothing\n\n if (!showDeleteConfirm) {\n setShowDeleteConfirm(true);\n } else {\n onDeleteChat(currentChatId);\n setIsOpen(false);\n setShowDeleteConfirm(false);\n }\n };\n\n const handleLockClick = () => {\n onLock();\n setIsOpen(false);\n };\n\n const handleResetClick = async () => {\n if (!showResetConfirm) {\n setShowResetConfirm(true);\n } else {\n const confirmed = confirm(\n \"⚠️ WARNING: This will permanently delete ALL chats and reset your password.\\n\\n\" +\n \"This action cannot be undone. Are you absolutely sure?\",\n );\n\n if (!confirmed) {\n setShowResetConfirm(false);\n return;\n }\n\n try {\n await clearMasterPassword();\n window.location.reload();\n } catch (error) {\n console.error(\"Reset failed:\", error);\n alert(\"Failed to reset: \" + (error as Error).message);\n }\n }\n };\n\n return (\n

\n setIsOpen(!isOpen)}\n className=\" w-8 h-8 pb-2 rounded-full font-bold bg-vibe-400 border border-transparent hover:border-vibe-250 text-vibe-100 hover:bg-vibe-500 transition-colors flex items-center justify-center shadow-lg\"\n title=\"Options\"\n >\n ...\n \n\n {isOpen && (\n

\n \n Lock\n \n \n {showDeleteConfirm ? \"⚠️ Click to confirm\" : \"Delete Chat\"}\n \n \n {showResetConfirm ? \"⚠️ Click to confirm\" : \"Reset Everything\"}\n \n

\n )}\n

\n );\n}\n",[],[194,241],{"_106":250,"_97":251,"_165":252,"_167":195,"_168":150,"_169":253,"_171":121,"_108":254,"_102":255,"_173":205,"_175":256,"_176":22,"_177":-5,"_178":257,"_100":258},"e15d0506-edb7-11f0-b9d9-42dde27851f2","StatusBadge.tsx","ca0120e4-d7b0-11f0-9689-42dde27851f2",1033,["D",1768003446408],7,"/** @jsxImportSource https://esm.sh/react */\nimport React from \"https://esm.sh/react\";\nimport type { WebLLMStatus } from \"../types.ts\";\n\ntype Props = {\n status: WebLLMStatus;\n};\n\nexport function StatusBadge({ status }: Props) {\n const styles = {\n idle: \"hidden\",\n loading: \"hidden\",\n ready: \"hidden\",\n error: \"bg-red-100 text-red-800\",\n };\n\n return (\n \n {status.state === \"loading\" && status.progress !== undefined\n ? `${status.progress}%`\n : status.message}\n \n );\n}\n",[],[194,251],{"_106":260,"_97":261,"_165":262,"_167":195,"_168":150,"_169":263,"_171":22,"_108":264,"_102":33,"_173":205,"_175":265,"_176":22,"_177":-5,"_178":266,"_100":267},"d3b8ff7c-008b-11f1-bd4c-42dde27851f2","ThinkingBubble.tsx","d3b90044-008b-11f1-bd4c-42dde27851f2",1151,["D",1770073597768],"/** @jsxImportSource https://esm.sh/react */\nimport React, { useEffect, useRef, useState } from \"https://esm.sh/react\";\n\ntype Props = {\n thinking: string;\n isStreaming: boolean;\n defaultExpanded?: boolean;\n};\n\nexport function ThinkingBubble({ thinking, isStreaming, defaultExpanded = false }: Props) {\n const [isExpanded, setIsExpanded] = useState(defaultExpanded || isStreaming);\n const contentRef = useRef(null);\n\n // Auto-expand while streaming\n useEffect(() => {\n if (isStreaming) {\n setIsExpanded(true);\n }\n }, [isStreaming]);\n\n // Auto-collapse after streaming completes (with delay)\n useEffect(() => {\n if (!isStreaming && thinking) {\n const timer = setTimeout(() => {\n setIsExpanded(false);\n }, 1000);\n return () => clearTimeout(timer);\n }\n }, [isStreaming, thinking]);\n\n // Generate preview text (first ~50 chars, ellipsed)\n const previewText = thinking.length > 50\n ? thinking.slice(0, 50).trim() + \"...\"\n : thinking;\n\n // Skip rendering if thinking is effectively empty\n if (!thinking || thinking.trim().length < 5) return null;\n\n return (\n

\n setIsExpanded(!isExpanded)}\n className=\"flex items-center gap-2 text-sm text-vibe-200 hover:text-vibe-100 transition-colors\"\n >\n {/* Chevron icon */}\n \n \n \n\n {/* Brain/thinking icon */}\n \n\n \n {isStreaming ? \"Thinking...\" : \"Thought process\"}\n \n\n {/* Preview text when collapsed */}\n {!isExpanded && (\n \n {previewText}\n \n )}\n \n\n {/* Expandable content */}\n \n \n {thinking}\n {isStreaming && (\n \n )}\n

\n \n \n );\n}\n",[],[194,261],[194],{"_106":270,"_97":271,"_165":272,"_167":-5,"_168":150,"_169":225,"_171":121,"_108":273,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":274,"_100":295},"c9db010c-d7b0-11f0-936d-42dde27851f2","hooks","c9db01d4-d7b0-11f0-936d-42dde27851f2",["D",1765581474921],[275,285],{"_106":276,"_97":277,"_165":278,"_167":272,"_168":150,"_169":279,"_171":121,"_108":280,"_102":281,"_173":205,"_175":282,"_176":22,"_177":-5,"_178":283,"_100":284},"4de4ff54-df84-11f0-9850-42dde27851f2","useEncryption.ts","40e70816-d9d9-11f0-a687-42dde27851f2",859,["D",1766441978393],3,"import { useState, useCallback } from \"https://esm.sh/react\";\nimport { encrypt, decrypt, type EncryptedPackage } from \"../lib/crypto.ts\";\nimport type { Message } from \"../types.ts\";\n\nexport function useEncryption() {\n const [password, setPassword] = useState(null);\n\n const encryptMessages = useCallback(async (messages: Message[]): Promise => {\n if (!password) throw new Error(\"No password set\");\n return encrypt(JSON.stringify(messages), password);\n }, [password]);\n\n const decryptMessages = useCallback(async (pkg: EncryptedPackage): Promise => {\n if (!password) throw new Error(\"No password set\");\n const json = await decrypt(pkg, password);\n return JSON.parse(json);\n }, [password]);\n\n const unlock = useCallback((pwd: string) => {\n setPassword(pwd);\n }, []);\n\n const lock = useCallback(() => {\n setPassword(null);\n }, []);\n\n const isUnlocked = password !== null;\n\n return {\n isUnlocked,\n unlock,\n lock,\n encryptMessages,\n decryptMessages\n };\n}",[],[271,277],{"_106":286,"_97":287,"_165":288,"_167":272,"_168":150,"_169":289,"_171":22,"_108":290,"_102":291,"_173":205,"_175":292,"_176":22,"_177":-5,"_178":293,"_100":294},"c6efc906-008b-11f1-8af1-42dde27851f2","useWebLLM.ts","ca180624-d7b0-11f0-895e-42dde27851f2",1150,["D",1770073576310],22,"/** @jsxImportSource https://esm.sh/react */\nimport { useCallback, useEffect, useRef, useState } from \"https://esm.sh/react\";\nimport type {\n Message,\n SendMessageOptions,\n SendMessageResult,\n WebLLMCompletionRequest,\n WebLLMCompletionResponse,\n WebLLMConfig,\n WebLLMEngine,\n WebLLMProgress,\n WebLLMStatus,\n WebLLMStreamChunk,\n} from \"../types.ts\";\nimport { getModelInfo } from \"../lib/models.ts\";\nimport { clearWebLLMCache, isCacheNetworkError } from \"../lib/webllm-cache.ts\";\n\n/**\n * Internal configuration constants for WebLLM hook.\n */\nconst WEBLLM_CONFIG: WebLLMConfig = {\n cdnUrl: \"https://cdn.jsdelivr.net/npm/@mlc-ai/web-llm@0.2.78/+esm\",\n systemPrompt:\n \"You are a helpful, friendly AI assistant. Keep responses concise, helpful and short. You can see the full conversation history and should reference previous messages when relevant.\",\n temperature: 0.7,\n maxTokens: 512,\n retryDelayMs: 1000,\n maxRetryAttempts: 1,\n} as const;\n\nconst THINKING_SYSTEM_PROMPT = `You are a helpful, friendly AI assistant. When answering questions that require reasoning, problem-solving, or complex explanations, first work through your thinking inside tags, then provide your final answer.\n\nFormat for complex questions:\n\n- Break down the problem\n- Consider different angles\n- Work through the logic step by step\n\n\nYour clear, helpful final answer here.\n\nFor simple greetings or straightforward factual questions, respond directly without thinking tags.`;\n\ntype ParserState = {\n phase: \"scanning\" | \"in_thinking\" | \"in_response\";\n thinkingBuffer: string;\n responseBuffer: string;\n tagBuffer: string;\n};\n\nfunction createThinkingParser() {\n let state: ParserState = {\n phase: \"scanning\",\n thinkingBuffer: \"\",\n responseBuffer: \"\",\n tagBuffer: \"\",\n };\n\n return {\n feed(chunk: string): { thinking: string; response: string; phase: \"thinking\" | \"response\" } {\n for (const char of chunk) {\n if (state.phase === \"scanning\") {\n state.tagBuffer += char;\n if (state.tagBuffer.endsWith(\"\")) {\n state.phase = \"in_thinking\";\n state.tagBuffer = \"\";\n } else if (state.tagBuffer.length > 15) {\n // No thinking tag found at start, treat as direct response\n state.responseBuffer += state.tagBuffer;\n state.tagBuffer = \"\";\n state.phase = \"in_response\";\n }\n } else if (state.phase === \"in_thinking\") {\n state.tagBuffer += char;\n if (state.tagBuffer.endsWith(\"\")) {\n state.thinkingBuffer += state.tagBuffer.slice(0, -11);\n state.tagBuffer = \"\";\n state.phase = \"in_response\";\n } else if (!state.tagBuffer.includes(\"<\")) {\n state.thinkingBuffer += state.tagBuffer;\n state.tagBuffer = \"\";\n }\n } else if (state.phase === \"in_response\") {\n state.responseBuffer += char;\n }\n }\n\n return {\n thinking: state.thinkingBuffer,\n response: state.responseBuffer,\n phase: state.phase === \"in_thinking\" ? \"thinking\" : \"response\",\n };\n },\n\n finalize(): { thinking: string; response: string } {\n // Handle incomplete tags gracefully\n if (state.phase === \"in_thinking\") {\n state.thinkingBuffer += state.tagBuffer;\n } else if (state.phase === \"scanning\") {\n state.responseBuffer += state.tagBuffer;\n } else {\n state.responseBuffer += state.tagBuffer;\n }\n return {\n thinking: state.thinkingBuffer.trim(),\n response: state.responseBuffer.trim(),\n };\n },\n\n reset() {\n state = {\n phase: \"scanning\",\n thinkingBuffer: \"\",\n responseBuffer: \"\",\n tagBuffer: \"\",\n };\n },\n };\n}\n\ninterface WebLLMModule {\n CreateMLCEngine(\n modelId: string,\n config: { initProgressCallback: WebLLMProgressCallback },\n ): Promise;\n}\n\n/**\n * React hook for managing WebLLM (browser-based LLM inference).\n *\n * Handles model loading, lifecycle management, and chat completion.\n * Models run entirely in the browser using WebGPU for privacy.\n *\n * @example\n * ```tsx\n * const { status, isReady, sendMessage } = useWebLLM(\"Phi-3-mini-4k-instruct-q4f16_1-MLC\");\n *\n * if (!isReady) {\n * return

{status.message} {status.progress}%

;\n * }\n *\n * const response = await sendMessage([{ role: \"user\", content: \"Hello!\" }]);\n * ```\n *\n * @param selectedModel - Model ID from AVAILABLE_MODELS\n * @returns Hook interface with status, isReady flag, and sendMessage function\n */\nexport function useWebLLM(selectedModel: string) {\n const [status, setStatus] = useState({\n state: \"idle\",\n message: \"Initializing...\",\n });\n const engineRef = useRef(null);\n const retryCountRef = useRef(0);\n const retryTimeoutRef = useRef(null);\n\n useEffect(() => {\n let cancelled = false;\n // Reset retry count when model changes\n retryCountRef.current = 0;\n\n async function loadModel() {\n const modelInfo = getModelInfo(selectedModel);\n const modelName = modelInfo ? modelInfo.name : selectedModel;\n\n console.log(\"[useWebLLM] Starting with model:\", selectedModel);\n\n if (!navigator.gpu) {\n setStatus({\n state: \"error\",\n message: \"WebGPU not supported - need Chrome/Edge\",\n });\n return;\n }\n\n try {\n // Dispose of previous engine if it exists\n if (engineRef.current) {\n console.log(\"[useWebLLM] Disposing previous engine...\");\n const prevEngine = engineRef.current;\n engineRef.current = null;\n\n try {\n await prevEngine.unload();\n } catch (disposeError) {\n console.warn(\n \"[useWebLLM] Error unloading previous engine:\",\n disposeError,\n );\n }\n }\n\n setStatus({\n state: \"loading\",\n message: \"Loading WebLLM library...\",\n progress: 0,\n });\n\n const webllm = await import(WEBLLM_CONFIG.cdnUrl) as WebLLMModule;\n\n if (cancelled) return;\n\n setStatus({\n state: \"loading\",\n message: `Downloading ${modelName}...`,\n progress: 0,\n });\n\n const engine = await webllm.CreateMLCEngine(selectedModel, {\n initProgressCallback: (progress: WebLLMProgress) => {\n if (cancelled) return;\n const pct = Math.round(progress.progress * 100);\n console.log(`[useWebLLM] Progress: ${pct}%`);\n setStatus({\n state: \"loading\",\n message: progress.text || `Downloading ${modelName}...`,\n progress: pct,\n });\n },\n });\n\n if (cancelled) return;\n\n console.log(\"[useWebLLM] Ready with\", modelName);\n engineRef.current = engine;\n setStatus({\n state: \"ready\",\n message: `${modelName} Ready`,\n progress: 100,\n });\n } catch (error: unknown) {\n const errorMessage = error instanceof Error\n ? error.message\n : \"Unknown error occurred\";\n console.error(\"[useWebLLM] Error:\", error);\n\n // If it's a cache network error and we haven't retried yet, clear cache and retry\n if (isCacheNetworkError(error) && retryCountRef.current === 0) {\n console.log(\n \"[useWebLLM] Detected cache corruption, clearing and retrying...\",\n );\n retryCountRef.current = 1;\n setStatus({\n state: \"loading\",\n message: \"Clearing corrupted cache and retrying...\",\n progress: 0,\n });\n\n await clearWebLLMCache();\n\n if (!cancelled) {\n // Retry loading after a short delay\n retryTimeoutRef.current = window.setTimeout(() => {\n if (!cancelled) loadModel();\n }, WEBLLM_CONFIG.retryDelayMs);\n }\n } else {\n setStatus({ state: \"error\", message: errorMessage });\n }\n }\n }\n\n loadModel();\n\n return () => {\n cancelled = true;\n if (retryTimeoutRef.current !== null) {\n clearTimeout(retryTimeoutRef.current);\n retryTimeoutRef.current = null;\n }\n };\n }, [selectedModel]);\n\n /**\n * Send messages to the LLM and get a response.\n * Automatically prepends system prompt to conversation.\n * Supports streaming and thinking mode.\n *\n * @param messages - Array of user/assistant messages\n * @param options - Optional settings for thinking mode and streaming\n * @returns Promise resolving to result with content and optional thinking\n * @throws Error if engine not ready or generation fails\n */\n const sendMessage = useCallback(\n async (messages: Message[], options?: SendMessageOptions): Promise => {\n console.log(\"[useWebLLM] Sending messages:\", messages);\n\n if (!engineRef.current) {\n throw new Error(\"Engine not ready\");\n }\n\n const enableThinking = options?.enableThinking ?? false;\n const onStream = options?.onStream;\n\n // Choose system prompt based on thinking mode\n const systemPrompt = enableThinking\n ? THINKING_SYSTEM_PROMPT\n : WEBLLM_CONFIG.systemPrompt;\n\n // Build messages array with system prompt\n const chatMessages = [\n { role: \"system\" as const, content: systemPrompt },\n ...messages.map((m) => ({ role: m.role, content: m.content })),\n ];\n\n console.log(\"[useWebLLM] Full chat context:\", chatMessages);\n\n // Use streaming if callback provided\n if (onStream) {\n const parser = createThinkingParser();\n\n try {\n const stream = await engineRef.current.chat.completions.create({\n messages: chatMessages,\n temperature: WEBLLM_CONFIG.temperature,\n max_tokens: enableThinking ? 1024 : WEBLLM_CONFIG.maxTokens,\n stream: true,\n }) as AsyncIterable;\n\n for await (const chunk of stream) {\n const deltaContent = chunk.choices[0]?.delta?.content || \"\";\n if (deltaContent) {\n const parsed = parser.feed(deltaContent);\n onStream({\n phase: parsed.phase,\n content: parsed.phase === \"thinking\" ? parsed.thinking : parsed.response,\n isComplete: false,\n });\n }\n }\n\n const final = parser.finalize();\n onStream({\n phase: \"response\",\n content: final.response,\n isComplete: true,\n });\n\n return {\n content: final.response,\n thinking: final.thinking || undefined,\n };\n } catch (streamError) {\n console.error(\"[useWebLLM] Stream error:\", streamError);\n const partial = parser.finalize();\n return {\n content: partial.response || \"Response interrupted. Please try again.\",\n thinking: partial.thinking || undefined,\n };\n }\n }\n\n // Non-streaming fallback\n const response = await engineRef.current.chat.completions.create({\n messages: chatMessages,\n temperature: WEBLLM_CONFIG.temperature,\n max_tokens: WEBLLM_CONFIG.maxTokens,\n }) as WebLLMCompletionResponse;\n\n console.log(\"[useWebLLM] Response:\", response);\n\n if (!response.choices || response.choices.length === 0) {\n throw new Error(\"No response choices returned from model\");\n }\n\n const content = response.choices[0]?.message?.content;\n if (!content) {\n throw new Error(\"Response message content is empty\");\n }\n\n // Parse thinking tags from complete response if thinking was enabled\n if (enableThinking) {\n const thinkingMatch = content.match(/([\\s\\S]*?)<\\/thinking>/);\n const thinking = thinkingMatch ? thinkingMatch[1].trim() : undefined;\n const cleanContent = content.replace(/[\\s\\S]*?<\\/thinking>/g, \"\").trim();\n return { content: cleanContent, thinking };\n }\n\n return { content };\n },\n [],\n );\n\n return { status, isReady: status.state === \"ready\", sendMessage };\n}\n",[],[271,287],[271],{"_106":297,"_97":298,"_165":299,"_167":-5,"_168":150,"_169":300,"_171":121,"_108":301,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":302,"_100":348},"e6c2eada-d9d8-11f0-9f30-42dde27851f2","lib","e6c2ebe8-d9d8-11f0-9f30-42dde27851f2",39,["D",1765818605605],[303,312,320,330,339],{"_106":304,"_97":305,"_165":306,"_167":299,"_168":150,"_169":184,"_171":121,"_108":307,"_102":308,"_173":205,"_175":309,"_176":22,"_177":-5,"_178":310,"_100":311},"d890f474-ecf3-11f0-b2da-42dde27851f2","api.ts","fe65e8a4-d9e7-11f0-b931-42dde27851f2",["D",1767919250293],18,"import type { EncryptedPackage, ChatListItem, MasterPasswordToken } from \"../types.ts\";\n\n// Get storage URL from window (injected by main.ts)\nfunction getStorageUrl(): string {\n return (window as any).STORAGE_URL || \"\";\n}\n\n// Get API key from window (injected by main.ts)\nfunction getApiKey(): string {\n return (window as any).STORAGE_API_KEY || \"\";\n}\n\n// Helper to create auth headers\nfunction getHeaders(includeContentType = false): HeadersInit {\n const headers: HeadersInit = {\n \"Authorization\": `Bearer ${getApiKey()}`\n };\n if (includeContentType) {\n headers[\"Content-Type\"] = \"application/json\";\n }\n return headers;\n}\n\nexport async function listChats(): Promise {\n const response = await fetch(`${getStorageUrl()}/list`, {\n headers: getHeaders()\n });\n if (!response.ok) throw new Error(\"Failed to load chat list\");\n const result = await response.json();\n return result.chats;\n}\n\nexport async function createChat(id: string, name: string, data: EncryptedPackage): Promise {\n const response = await fetch(`${getStorageUrl()}/create`, {\n method: \"POST\",\n headers: getHeaders(true),\n body: JSON.stringify({ id, name, data })\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Create failed\");\n }\n}\n\nexport async function saveChat(id: string, data: EncryptedPackage): Promise {\n const response = await fetch(`${getStorageUrl()}/save`, {\n method: \"POST\",\n headers: getHeaders(true),\n body: JSON.stringify({ id, data })\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Save failed\");\n }\n}\n\nexport async function renameChat(id: string, name: string): Promise {\n const response = await fetch(`${getStorageUrl()}/rename`, {\n method: \"POST\",\n headers: getHeaders(true),\n body: JSON.stringify({ id, name })\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Rename failed\");\n }\n}\n\nexport async function loadChat(id: string): Promise {\n const response = await fetch(`${getStorageUrl()}/load?id=${encodeURIComponent(id)}`, {\n headers: getHeaders()\n });\n if (response.status === 404) return null;\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Load failed\");\n }\n const result = await response.json();\n return result.data;\n}\n\nexport async function deleteChat(id: string): Promise {\n const response = await fetch(`${getStorageUrl()}/delete?id=${encodeURIComponent(id)}`, {\n method: \"DELETE\",\n headers: getHeaders()\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Delete failed\");\n }\n}\n\nexport async function deleteAllChats(): Promise {\n const response = await fetch(`${getStorageUrl()}/delete-all`, {\n method: \"DELETE\",\n headers: getHeaders()\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Delete all failed\");\n }\n}\n\nexport async function checkPasswordStatus(): Promise {\n const response = await fetch(`${getStorageUrl()}/config/password-status`, {\n headers: getHeaders()\n });\n if (!response.ok) throw new Error(\"Failed to check password status\");\n const result = await response.json();\n return result.hasPassword;\n}\n\nexport async function savePasswordToken(token: MasterPasswordToken): Promise {\n const response = await fetch(`${getStorageUrl()}/config/save-password-token`, {\n method: \"POST\",\n headers: getHeaders(true),\n body: JSON.stringify({ token })\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Failed to save password token\");\n }\n}\n\nexport async function getPasswordToken(): Promise {\n const response = await fetch(`${getStorageUrl()}/config/password-token`, {\n headers: getHeaders()\n });\n if (!response.ok) throw new Error(\"Failed to get password token\");\n const result = await response.json();\n return result.token;\n}\n\nexport async function resetAllData(): Promise {\n const response = await fetch(`${getStorageUrl()}/config/reset-all`, {\n method: \"POST\",\n headers: getHeaders(true)\n });\n if (!response.ok) {\n const err = await response.json();\n throw new Error(err.error || \"Reset failed\");\n }\n}",[],[298,305],{"_106":313,"_97":314,"_165":315,"_167":299,"_168":150,"_169":184,"_171":121,"_108":316,"_102":281,"_173":205,"_175":317,"_176":22,"_177":-5,"_178":318,"_100":319},"d891073e-ecf3-11f0-b2da-42dde27851f2","crypto.ts","28dbac9a-d9d9-11f0-afba-42dde27851f2",["D",1767919250293],"// Encode text to bytes\nfunction textToBytes(text: string): Uint8Array {\n return new TextEncoder().encode(text);\n}\n\n// Decode bytes to text\nfunction bytesToText(bytes: ArrayBuffer): string {\n return new TextDecoder().decode(bytes);\n}\n\n// Derive encryption key from password + salt\nasync function deriveKey(password: string, salt: Uint8Array): Promise {\n const keyMaterial = await crypto.subtle.importKey(\n \"raw\",\n textToBytes(password),\n \"PBKDF2\",\n false,\n [\"deriveKey\"]\n );\n\n return crypto.subtle.deriveKey(\n {\n name: \"PBKDF2\",\n salt,\n iterations: 100000,\n hash: \"SHA-256\"\n },\n keyMaterial,\n { name: \"AES-GCM\", length: 256 },\n false,\n [\"encrypt\", \"decrypt\"]\n );\n}\n\n// Encrypted package type\nexport type EncryptedPackage = {\n salt: number[];\n iv: number[];\n ciphertext: number[];\n};\n\n// Encrypt text with password\nexport async function encrypt(text: string, password: string): Promise {\n const salt = crypto.getRandomValues(new Uint8Array(16));\n const iv = crypto.getRandomValues(new Uint8Array(12));\n \n // Derive key from password FIRST\n const key = await deriveKey(password, salt);\n\n const encryptedBytes = await crypto.subtle.encrypt(\n { name: \"AES-GCM\", iv },\n key, // This must be a CryptoKey, not a string\n textToBytes(text)\n );\n\n return {\n salt: Array.from(salt),\n iv: Array.from(iv),\n ciphertext: Array.from(new Uint8Array(encryptedBytes))\n };\n}\n\n// Decrypt with password\nexport async function decrypt(pkg: EncryptedPackage, password: string): Promise {\n const salt = new Uint8Array(pkg.salt);\n const iv = new Uint8Array(pkg.iv);\n const ciphertext = new Uint8Array(pkg.ciphertext);\n\n // Derive the same key from password + stored salt\n const key = await deriveKey(password, salt);\n\n const decryptedBytes = await crypto.subtle.decrypt(\n { name: \"AES-GCM\", iv },\n key,\n ciphertext\n );\n\n return bytesToText(decryptedBytes);\n}\n\n// Import API functions for database storage\nimport { savePasswordToken, getPasswordToken, checkPasswordStatus, resetAllData } from \"./api.ts\";\n\n// Master password verification\nconst PASSWORD_VERIFICATION_STRING = \"LOCALS_AI_VALID_PASSWORD_v1\";\n\nexport type MasterPasswordToken = {\n version: number;\n verificationData: EncryptedPackage;\n createdAt: string;\n};\n\n// Create verification token for master password\nexport async function createMasterPasswordToken(password: string): Promise {\n const verificationData = await encrypt(PASSWORD_VERIFICATION_STRING, password);\n const token: MasterPasswordToken = {\n version: 1,\n verificationData,\n createdAt: new Date().toISOString()\n };\n\n // Save to database (source of truth)\n await savePasswordToken(token);\n}\n\n// Verify master password against stored token\nexport async function verifyMasterPassword(password: string): Promise {\n try {\n const token = await getPasswordToken();\n if (!token) return false;\n\n const decrypted = await decrypt(token.verificationData, password);\n return decrypted === PASSWORD_VERIFICATION_STRING;\n } catch {\n return false; // Decryption failed = wrong password\n }\n}\n\n// Check if master password has been set\nexport async function hasMasterPassword(): Promise {\n try {\n return await checkPasswordStatus();\n } catch (error) {\n console.error(\"Failed to check password status:\", error);\n return false;\n }\n}\n\n// Clear master password and reset all data\nexport async function clearMasterPassword(): Promise {\n await resetAllData();\n // Also clear sessionStorage\n sessionStorage.clear();\n}",[],[298,314],{"_106":321,"_97":322,"_165":323,"_167":299,"_168":150,"_169":324,"_171":121,"_108":325,"_102":326,"_173":205,"_175":327,"_176":22,"_177":-5,"_178":328,"_100":329},"605b87b6-edb7-11f0-91ff-42dde27851f2","deviceDetection.ts","46619566-edb3-11f0-8e47-42dde27851f2",1027,["D",1768003229975],1,"/**\n * Device and capability detection utilities for determining browser support\n */\n\n/**\n * Detects if the current device is a mobile device based on user agent\n * @returns true if mobile device detected, false otherwise\n */\nexport function isMobileDevice(): boolean {\n if (typeof navigator === \"undefined\") {\n return false;\n }\n\n // Check for common mobile device indicators in user agent\n const mobileRegex = /Mobile|Android|iPhone|iPad|iPod|BlackBerry|Windows Phone/i;\n const isMobileUA = mobileRegex.test(navigator.userAgent);\n\n // Special check for iPad (modern iPads report as Mac in user agent)\n const isMacWithTouch = /Macintosh/i.test(navigator.userAgent) &&\n navigator.maxTouchPoints > 1;\n\n return isMobileUA || isMacWithTouch;\n}\n\n/**\n * Checks if the browser supports WebGPU (synchronous check only)\n * Note: This only checks if navigator.gpu exists, not if it's functional\n * @returns true if WebGPU API is present, false otherwise\n */\nexport function hasWebGPUSupport(): boolean {\n // SSR-safe check following pattern from MasterPasswordScreen.tsx:24\n return typeof navigator !== \"undefined\" && !!navigator.gpu;\n}\n\n/**\n * Tests if WebGPU is actually functional (not just present but blocked)\n * @returns Promise that resolves to true if WebGPU works, false otherwise\n */\nexport async function isWebGPUFunctional(): Promise {\n if (!hasWebGPUSupport()) {\n return false;\n }\n\n try {\n // Try to actually request an adapter - this will fail if WebGPU is blocked\n const adapter = await navigator.gpu.requestAdapter();\n return adapter !== null;\n } catch (error) {\n // If requestAdapter throws or fails, WebGPU is not functional\n return false;\n }\n}\n\n/**\n * Determines if the current environment is unsupported for the chat application\n * @returns true if environment is unsupported (mobile OR no WebGPU), false otherwise\n */\nexport function isUnsupportedEnvironment(): boolean {\n return isMobileDevice() || !hasWebGPUSupport();\n}\n",[],[298,322],{"_106":331,"_97":332,"_165":333,"_167":299,"_168":150,"_169":334,"_171":121,"_108":335,"_102":33,"_173":205,"_175":336,"_176":22,"_177":-5,"_178":337,"_100":338},"b89039a8-dd46-11f0-be05-42dde27851f2","models.ts","b8903a7a-dd46-11f0-be05-42dde27851f2",686,["D",1766195626235],"export type ModelSize = \"light\" | \"medium\" | \"large\";\n\nexport interface ModelInfo {\n id: string;\n name: string;\n size: ModelSize;\n downloadSize: string;\n description: string;\n}\n\nexport const AVAILABLE_MODELS: ModelInfo[] = [\n {\n id: \"Phi-3-mini-4k-instruct-q4f16_1-MLC\",\n name: \"Phi-3 Mini\",\n size: \"light\",\n downloadSize: \"2.3GB\",\n description: \"Fast and efficient\",\n },\n {\n id: \"Llama-3.2-3B-Instruct-q4f16_1-MLC\",\n name: \"Llama 3.2 3B\",\n size: \"medium\",\n downloadSize: \"1.9GB\",\n description: \"Balanced performance\",\n },\n {\n id: \"Llama-3.1-8B-Instruct-q4f32_1-MLC\",\n name: \"Llama 3.1 8B\",\n size: \"large\",\n downloadSize: \"4.7GB\",\n description: \"Best quality\",\n },\n];\n\nexport function getModelInfo(modelId: string): ModelInfo | undefined {\n return AVAILABLE_MODELS.find((model) => model.id === modelId);\n}\n\nexport function getModelSizeLabel(size: ModelSize): string {\n const labels = {\n light: \"Light\",\n medium: \"Medium\",\n large: \"Large\",\n };\n return labels[size];\n}\n",[],[298,332],{"_106":340,"_97":341,"_165":342,"_167":299,"_168":150,"_169":343,"_171":121,"_108":344,"_102":33,"_173":205,"_175":345,"_176":22,"_177":-5,"_178":346,"_100":347},"dd7d1a14-f002-11f0-8b62-42dde27851f2","webllm-cache.ts","dd7d1ae6-f002-11f0-8b62-42dde27851f2",1065,["D",1768255554499],"/**\n * Cache management utilities for WebLLM model downloads.\n *\n * WebLLM caches downloaded model weights in browser Cache API.\n * These utilities detect and clear corrupted caches to enable retry.\n *\n * @module lib/webllm-cache\n */\n\nconst WEBLLM_CACHE_PATTERNS = ['webllm', 'web-llm', 'mlc'] as const;\n\n/**\n * Clear WebLLM-related caches to fix corrupted downloads.\n *\n * @returns Promise that resolves when cache clearing is complete\n */\nexport async function clearWebLLMCache(): Promise {\n try {\n const cacheNames = await caches.keys();\n const webllmCaches = cacheNames.filter(name =>\n WEBLLM_CACHE_PATTERNS.some(pattern => name.includes(pattern))\n );\n\n console.log(\"[webllm-cache] Clearing corrupted caches:\", webllmCaches);\n await Promise.all(webllmCaches.map(name => caches.delete(name)));\n\n if (webllmCaches.length === 0) {\n console.log(\"[webllm-cache] Clearing all caches to fix corruption\");\n await Promise.all(cacheNames.map(name => caches.delete(name)));\n }\n } catch (error) {\n console.warn(\"[webllm-cache] Error clearing cache:\", error);\n }\n}\n\n/**\n * Check if an error is a cache-related network error.\n *\n * @param error - The error object to check\n * @returns true if error is cache-related network error\n */\nexport function isCacheNetworkError(error: unknown): boolean {\n if (!(error instanceof Error)) return false;\n\n const message = error.message.toLowerCase();\n return message.includes('cache') &&\n (message.includes('network') || message.includes('content_length_mismatch'));\n}\n",[],[298,341],[298],{"_106":350,"_97":101,"_165":351,"_167":-5,"_168":150,"_169":352,"_171":121,"_108":353,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":354,"_100":384},"c9bb7d64-d7b0-11f0-a0a6-42dde27851f2","c9bb7e2c-d7b0-11f0-a0a6-42dde27851f2",11,["D",1765581474705],[355,364,374],{"_106":356,"_97":357,"_165":358,"_167":351,"_168":150,"_169":103,"_171":22,"_108":359,"_102":360,"_173":205,"_175":361,"_176":22,"_177":-5,"_178":362,"_100":363},"f2f1100a-008b-11f1-9b51-42dde27851f2","MainScreen.tsx","3b419c04-dc6e-11f0-9833-42dde27851f2",["D",1770073650143],393,"/** @jsxImportSource https://esm.sh/react */\nimport React, { useEffect, useRef, useState } from \"https://esm.sh/react\";\nimport type {\n ChatListItem,\n EncryptedPackage,\n Message,\n SendMessageOptions,\n SendMessageResult,\n WebLLMStatus,\n} from \"../types.ts\";\nimport {\n createChat,\n deleteAllChats,\n deleteChat,\n listChats,\n loadChat,\n renameChat,\n saveChat,\n} from \"../lib/api.ts\";\nimport { ChatNameModal } from \"../components/ChatNameModal.tsx\";\nimport { OptionsMenu } from \"../components/OptionsMenu.tsx\";\nimport { StatusBadge } from \"../components/StatusBadge.tsx\";\nimport { MessageBubble } from \"../components/MessageBubble.tsx\";\nimport { EmptyState } from \"../components/EmptyState.tsx\";\nimport { ModelSelector } from \"../components/ModelSelector.tsx\";\n\ntype Props = {\n status: WebLLMStatus;\n isReady: boolean;\n onSendMessage: (messages: Message[], options?: SendMessageOptions) => Promise;\n onEncryptMessages: (messages: Message[]) => Promise;\n onDecryptMessages: (data: EncryptedPackage) => Promise;\n onLock: () => void;\n selectedModel: string;\n onModelChange: (modelId: string) => void;\n};\n\nfunction generateChatId(): string {\n const array = new Uint8Array(8);\n crypto.getRandomValues(array);\n return Array.from(array).map((b) => b.toString(16).padStart(2, \"0\")).join(\"\");\n}\n\nexport function MainScreen({\n status,\n isReady,\n onSendMessage,\n onEncryptMessages,\n onDecryptMessages,\n onLock,\n selectedModel,\n onModelChange,\n}: Props) {\n const [chats, setChats] = useState([]);\n const [selectedChat, setSelectedChat] = useState(null);\n const [messages, setMessages] = useState([]);\n const [input, setInput] = useState(\"\");\n const [isGenerating, setIsGenerating] = useState(false);\n const [saveStatus, setSaveStatus] = useState(null);\n const [loading, setLoading] = useState(true);\n const [error, setError] = useState(null);\n\n // Thinking mode state\n const [thinkingEnabled, setThinkingEnabled] = useState(false);\n const [streamingMessage, setStreamingMessage] = useState<{\n thinking: string;\n content: string;\n phase: \"thinking\" | \"response\";\n } | null>(null);\n\n // Modal states\n const [isModalOpen, setIsModalOpen] = useState(false);\n const [modalMode, setModalMode] = useState<\"create\" | \"rename\">(\"create\");\n const [modalLoading, setModalLoading] = useState(false);\n const [modalError, setModalError] = useState(null);\n\n // Mobile menu state\n const [isMobileMenuOpen, setIsMobileMenuOpen] = useState(false);\n\n const messagesEndRef = useRef(null);\n\n // Load chats on mount\n useEffect(() => {\n async function fetchChats() {\n try {\n const list = await listChats();\n setChats(list);\n\n // Auto-select the most recent chat\n if (list.length > 0) {\n await handleSelectChat(list[0]);\n }\n } catch (e) {\n console.error(\"Failed to load chats:\", e);\n setError(\"Failed to load chats\");\n } finally {\n setLoading(false);\n }\n }\n fetchChats();\n }, []);\n\n // Auto-scroll to bottom when messages change\n useEffect(() => {\n messagesEndRef.current?.scrollIntoView({ behavior: \"smooth\" });\n }, [messages]);\n\n const handleSelectChat = async (chat: ChatListItem) => {\n setError(null);\n setLoading(true);\n setIsMobileMenuOpen(false); // Close mobile menu when chat is selected\n\n try {\n const existingData = await loadChat(chat.id);\n\n if (!existingData) {\n setError(\"Chat not found\");\n setLoading(false);\n return;\n }\n\n const decryptedMessages = await onDecryptMessages(existingData);\n setMessages(decryptedMessages);\n setSelectedChat(chat);\n } catch (e: any) {\n setError(\"Failed to decrypt chat: \" + e.message);\n }\n\n setLoading(false);\n };\n\n const handleNewChat = () => {\n setModalMode(\"create\");\n setModalError(null);\n setIsModalOpen(true);\n };\n\n const handleRenameChat = () => {\n if (!selectedChat) return;\n setModalMode(\"rename\");\n setModalError(null);\n setIsModalOpen(true);\n };\n\n const handleModalSave = async (name: string) => {\n setModalLoading(true);\n setModalError(null);\n\n try {\n if (modalMode === \"create\") {\n const id = generateChatId();\n const encrypted = await onEncryptMessages([]);\n await createChat(id, name, encrypted);\n\n const newChat: ChatListItem = {\n id,\n name,\n createdAt: new Date().toISOString(),\n updatedAt: new Date().toISOString(),\n };\n\n setChats([newChat, ...chats]);\n setSelectedChat(newChat);\n setMessages([]);\n setIsModalOpen(false);\n } else if (modalMode === \"rename\" && selectedChat) {\n await renameChat(selectedChat.id, name);\n\n const updatedChat = { ...selectedChat, name };\n setSelectedChat(updatedChat);\n setChats(chats.map((c) => c.id === selectedChat.id ? updatedChat : c));\n setIsModalOpen(false);\n }\n } catch (e: any) {\n setModalError(e.message);\n }\n\n setModalLoading(false);\n };\n\n const handleDeleteData = async () => {\n try {\n await deleteAllChats();\n setChats([]);\n setSelectedChat(null);\n setMessages([]);\n } catch (e: any) {\n console.error(\"Failed to delete all chats:\", e);\n alert(\"Failed to delete chats: \" + e.message);\n }\n };\n\n const handleDeleteChat = async (chatId: string) => {\n try {\n await deleteChat(chatId);\n\n // Remove from chat list\n const updatedChats = chats.filter((c) => c.id !== chatId);\n setChats(updatedChats);\n\n // If deleted chat was selected, clear selection\n if (selectedChat?.id === chatId) {\n setSelectedChat(null);\n setMessages([]);\n\n // Auto-select another chat if available\n if (updatedChats.length > 0) {\n await handleSelectChat(updatedChats[0]);\n }\n }\n } catch (e: any) {\n console.error(\"Failed to delete chat:\", e);\n alert(\"Failed to delete chat: \" + e.message);\n }\n };\n\n const handleSend = async () => {\n if (!input.trim() || !isReady || isGenerating || !selectedChat) return;\n\n const userMessage: Message = { role: \"user\", content: input.trim() };\n const newMessages = [...messages, userMessage];\n\n setMessages(newMessages);\n setInput(\"\");\n setIsGenerating(true);\n setStreamingMessage({ thinking: \"\", content: \"\", phase: \"thinking\" });\n\n try {\n const result = await onSendMessage(newMessages, {\n enableThinking: thinkingEnabled,\n onStream: (update) => {\n setStreamingMessage((prev) => {\n if (!prev) return { thinking: \"\", content: \"\", phase: update.phase };\n return {\n thinking: update.phase === \"thinking\" ? update.content : prev.thinking,\n content: update.phase === \"response\" ? update.content : prev.content,\n phase: update.phase,\n };\n });\n },\n });\n\n const assistantMessage: Message = {\n role: \"assistant\",\n content: result.content,\n thinking: result.thinking,\n };\n\n const updatedMessages = [...newMessages, assistantMessage];\n setMessages(updatedMessages);\n await saveConversation(updatedMessages);\n } catch (err: any) {\n console.error(\"Chat error:\", err);\n setMessages([...newMessages, {\n role: \"assistant\",\n content: `Error: ${err.message}`,\n }]);\n }\n\n setStreamingMessage(null);\n setIsGenerating(false);\n };\n\n const saveConversation = async (msgs: Message[]) => {\n if (!selectedChat) return;\n\n try {\n setSaveStatus(\"Saving...\");\n const encrypted = await onEncryptMessages(msgs);\n await saveChat(selectedChat.id, encrypted);\n setSaveStatus(\"Saved ✓\");\n setTimeout(() => setSaveStatus(null), 2000);\n\n // Update the chat's updatedAt timestamp in the list\n const updatedChats = chats.map((c) =>\n c.id === selectedChat.id\n ? { ...c, updatedAt: new Date().toISOString() }\n : c\n );\n setChats(\n updatedChats.sort((a, b) =>\n new Date(b.updatedAt).getTime() - new Date(a.updatedAt).getTime()\n ),\n );\n } catch (err: any) {\n console.error(\"Save error:\", err);\n setSaveStatus(\"Save failed\");\n }\n };\n\n const handleKeyDown = (e: React.KeyboardEvent) => {\n if (e.key === \"Enter\" && !e.shiftKey) {\n e.preventDefault();\n handleSend();\n }\n };\n\n return (\n

\n {/* Left Panel - Conversation List */}\n \n

\n Locals AI\n

\n \n \n +\n {\" \"}\n New Chat\n \n

\n\n

\n {loading && chats.length === 0\n ? (\n

\n Loading...\n

\n )\n : chats.length === 0\n ? (\n

No chats yet

\n )\n : (\n

\n Chats\n

\n {chats.map((chat) => (\n handleSelectChat(chat)}\n className={`w-[calc(100%-10px)] text-left px-4 py-1 h-fit hover:bg-vibe-300 hover:text-vibe-100 rounded-lg ${\n selectedChat?.id === chat.id\n ? \"text-white\"\n : \"text-vibe-200 \"\n }`}\n >\n

\n {chat.name}\n

\n \n ))}\n

\n )}\n

\n\n {/* Mobile overlay */}\n {isMobileMenuOpen && (\n setIsMobileMenuOpen(false)}\n />\n )}\n\n {/* Right Panel - Chat View */}\n

\n {selectedChat\n ? (\n <>\n {/* Chat Header */}\n

\n {/* Hamburger menu for mobile */}\n setIsMobileMenuOpen(!isMobileMenuOpen)}\n className=\"md:hidden p-2 text-white hover:bg-vibe-500 rounded-lg min-h-[44px] min-w-[44px]\"\n aria-label=\"Toggle menu\"\n >\n \n \n \n \n

\n \n {selectedChat.name}\n \n

\n {new Date(selectedChat.updatedAt)\n .toLocaleDateString()}\n

\n {saveStatus && (\n \n {saveStatus}\n \n )}\n \n\n \n

\n\n {/* Messages */}\n

\n {messages.length === 0 && }\n {messages.map((m, i) => (\n \n ))}\n {isGenerating && (\n

\n Thinking...\n

\n )}\n

\n\n {/* Input */}\n \n \n )\n : (\n

\n \n \n +\n {\" \"}\n New Chat\n \n

\n )}\n

\n\n {/* Chat Name Modal */}\n setIsModalOpen(false)}\n isLoading={modalLoading}\n error={modalError}\n />\n \n );\n}\n",[],[101,357],{"_106":365,"_97":366,"_165":367,"_167":351,"_168":150,"_169":368,"_171":121,"_108":369,"_102":370,"_173":205,"_175":371,"_176":22,"_177":-5,"_178":372,"_100":373},"b6aa8db6-f013-11f0-9008-42dde27851f2","MasterPasswordScreen.tsx","443bb97e-dc65-11f0-9c23-42dde27851f2",1096,["D",1768262790810],208,"/** @jsxImportSource https://esm.sh/react */\nimport React, { useState } from \"https://esm.sh/react\";\nimport type { MasterPasswordMode } from \"../types.ts\";\n\ntype Props = {\n mode: MasterPasswordMode;\n onUnlock: (password: string) => void;\n onCreate: (password: string) => void;\n error?: string | null;\n isLoading?: boolean;\n};\n\nexport function MasterPasswordScreen({\n mode,\n onUnlock,\n onCreate,\n error,\n isLoading,\n}: Props) {\n const [password, setPassword] = useState(\"\");\n const [validationError, setValidationError] = useState(\"\");\n\n // Check for WebGPU support\n const hasWebGPU = typeof navigator !== \"undefined\" && !!navigator.gpu;\n\n const handleSubmit = (e: React.FormEvent) => {\n e.preventDefault();\n\n if (password.length < 6) {\n setValidationError(\"Minimum 6 characters\");\n return;\n }\n\n setValidationError(\"\");\n if (mode === \"create\") {\n onCreate(password);\n } else {\n onUnlock(password);\n }\n };\n\n const displayError = error || validationError;\n\n return (\n

\n Locals AI\n

\n Learn More\n

\n Truly private processing in your browser\n

\n {!hasWebGPU && (\n

\n WebGPU not supported - need Chrome/Edge\n

\n This app requires WebGPU to run AI models in your browser.\n Please use Chrome or Edge.\n

\n )}\n

\n\n

\n \n {mode === \"create\"\n ? \"Create a conversation password to continue\"\n : \"Unlock your encrypted conversations\"}\n \n

\n {displayError && (\n

{displayError}

\n )}\n

\n {\n setPassword(e.target.value);\n setValidationError(\"\");\n }}\n placeholder=\"Minimum 6 characters\"\n disabled={isLoading}\n className=\"px-3 sm:px-4 py-3 min-h-[44px] text-white placeholder-[#99978F] bg-transparent focus:outline-none focus:ring-1 focus:ring-transparent\"\n autoFocus\n />\n \n {isLoading\n ? \"Processing...\"\n : mode === \"create\"\n ? \"Start\"\n : \"Unlock\"}\n \n

\n {mode === \"create\"\n ? \"If you forget this password, your data will be permanently lost. There is no recovery option.\"\n : \"Your password never leaves your device. It's used to decrypt your conversations locally.\"}\n

\n \n Make this repo yours →\n \n

\n );\n}\n",[],[101,366],{"_106":375,"_97":376,"_165":377,"_167":351,"_168":150,"_169":378,"_171":121,"_108":379,"_102":380,"_173":205,"_175":381,"_176":22,"_177":-5,"_178":382,"_100":383},"19f37bda-edb7-11f0-9a29-42dde27851f2","UnsupportedScreen.tsx","547192e6-edb3-11f0-8f2b-42dde27851f2",1026,["D",1768003111860],16,"/** @jsxImportSource https://esm.sh/react */\nimport React from \"https://esm.sh/react\";\n\ntype Props = {\n reason?: \"mobile\" | \"no-webgpu\";\n};\n\nexport function UnsupportedScreen({ reason = \"mobile\" }: Props) {\n const isMobile = reason === \"mobile\";\n return (\n

\n Locals AI\n

\n \n Learn More\n \n

\n Truly private processing in your browser\n

\n\n {/* Main warning box */}\n

\n \n \n \n

\n {isMobile\n ? \"Mobile Browsers Not Supported\"\n : \"Browser Not Supported\"}\n

\n {isMobile\n ? \"This application requires WebGPU technology to run AI models locally in your browser. WebGPU is not currently available on mobile devices.\"\n : \"This application requires WebGPU technology to run AI models locally in your browser. Your current browser does not support WebGPU.\"}\n

\n {isMobile\n ? \"Please access this application from a desktop computer using:\"\n : \"Please use one of the following browsers:\"}\n

Chrome (version 113 or later)
Edge (version 113 or later)

\n Note: Firefox and Safari support is not yet available\n

\n\n {/* Privacy reminder */}\n

\n Privacy First:{\" \"}\n Your conversations never leave your device. All AI processing\n happens locally in your browser.\n

\n );\n}\n",[],[101,376],[101],{"_106":386,"_97":387,"_165":388,"_167":-5,"_168":150,"_169":389,"_171":121,"_108":390,"_102":281,"_173":391,"_175":392,"_176":22,"_177":-5,"_178":393,"_100":394},"93affb6a-ea6a-11f0-9b9a-42dde27851f2",".vtignore","1eb284c0-d7ac-11f0-ba34-42dde27851f2",874,["D",1767640391361],"file",".git\n.vscode\n.cursorrules\n.DS_Store\nnode_modules\nvendor\n.claude\nclaude",[],[387],{"_106":396,"_97":397,"_165":398,"_167":-5,"_168":150,"_169":399,"_171":121,"_108":400,"_102":33,"_173":391,"_175":401,"_176":22,"_177":-5,"_178":402,"_100":403},"1eae19ee-d7ac-11f0-b459-42dde27851f2","AGENTS.md","1eae1aa2-d7ac-11f0-b459-42dde27851f2",2,["D",1765579469717],"You are an advanced assistant specialized in generating Val Town code.\n\n## Core Guidelines\n\n- Ask clarifying questions when requirements are ambiguous\n- Provide complete, functional solutions rather than skeleton implementations\n- Test your logic against edge cases before presenting the final solution\n- Ensure all code follows Val Town's specific platform requirements\n- If a section of code that you're working on is getting too complex, consider refactoring it into subcomponents\n\n## Code Standards\n\n- Generate code in TypeScript or TSX\n- Add appropriate TypeScript types and interfaces for all data structures\n- Prefer official SDKs or libraries than writing API calls directly\n- Ask the user to supply API or library documentation if you are at all unsure about it\n- **Never bake in secrets into the code** - always use environment variables\n- Include comments explaining complex logic (avoid commenting obvious operations)\n- Follow modern ES6+ conventions and functional programming practices if possible\n\n## Types of triggers\n\n### 1. HTTP Trigger\n\n- Create web APIs and endpoints\n- Handle HTTP requests and responses\n- Example structure:\n\n```ts\nexport default async function (req: Request) {\n return new Response(\"Hello World\");\n}\n```\n\nFiles that are HTTP triggers have http in their name like `foobar.http.tsx`\n\n### 2. Cron Triggers\n\n- Run on a schedule\n- Use cron expressions for timing\n- Example structure:\n\n```ts\nexport default async function () {\n // Scheduled task code\n}\n```\n\nFiles that are Cron triggers have cron in their name like `foobar.cron.tsx`\n\n### 3. Email Triggers\n\n- Process incoming emails\n- Handle email-based workflows\n- Example structure:\n\n```ts\nexport default async function (email: Email) {\n // Process email\n}\n```\n\nFiles that are Email triggers have email in their name like `foobar.email.tsx`\n\n\n## Val Town Standard Libraries\n\nVal Town provides several hosted services and utility functions.\n\n### Blob Storage\n\n```ts\nimport { blob } from \"https://esm.town/v/std/blob\";\nawait blob.setJSON(\"myKey\", { hello: \"world\" });\nlet blobDemo = await blob.getJSON(\"myKey\");\nlet appKeys = await blob.list(\"app_\");\nawait blob.delete(\"myKey\");\n```\n\n### SQLite\n\n```ts\nimport { sqlite } from \"https://esm.town/v/stevekrouse/sqlite\";\nconst TABLE_NAME = 'todo_app_users_2';\n// Create table - do this before usage and change table name when modifying schema\nawait sqlite.execute(`CREATE TABLE IF NOT EXISTS ${TABLE_NAME} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n name TEXT NOT NULL\n)`);\n// Query data\nconst result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} WHERE id = ?`, [1]);\n```\n\nNote: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.\n\n### OpenAI\n\n```ts\nimport { OpenAI } from \"https://esm.town/v/std/openai\";\nconst openai = new OpenAI();\nconst completion = await openai.chat.completions.create({\n messages: [\n { role: \"user\", content: \"Say hello in a creative way\" },\n ],\n model: \"gpt-4o-mini\",\n max_tokens: 30,\n});\n```\n\n### Email\n\n```ts\nimport { email } from \"https://esm.town/v/std/email\";\n// By default emails the owner of the val\nawait email({ \n subject: \"Hi\", \n text: \"Hi\", \n html: \"

Hi

\"\n});\n```\n\n## Val Town Utility Functions\n\nVal Town provides several utility functions to help with common project tasks.\n\n### Importing Utilities\n\nAlways import utilities with version pins to avoid breaking changes:\n\n```ts\nimport { parseProject, readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n```\n\n### Available Utilities\n\n\n#### **serveFile** - Serve project files with proper content types\n\nFor example, in Hono:\n\n```ts\n// serve all files in frontend/ and shared/\napp.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\napp.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n```\n\n#### **readFile** - Read files from within the project:\n\n```ts\n// Read a file from the project\nconst fileContent = await readFile(\"/frontend/index.html\", import.meta.url);\n```\n\n#### **listFiles** - List all files in the project\n\n```ts\nconst files = await listFiles(import.meta.url);\n```\n\n#### **parseProject** - Extract information about the current project from import.meta.url\n\nThis is useful for including info for linking back to a val, ie in \"view source\" urls:\n\n```ts\nconst projectVal = parseProject(import.meta.url);\nconsole.log(projectVal.username); // Owner of the project\nconsole.log(projectVal.name); // Project name\nconsole.log(projectVal.version); // Version number\nconsole.log(projectVal.branch); // Branch name\nconsole.log(projectVal.links.self.project); // URL to the project page\n```\n\nHowever, it's *extremely importing* to note that `parseProject` and other Standard Library utilities ONLY RUN ON THE SERVER.\nIf you need access to this data on the client, run it in the server and pass it to the client by splicing it into the HTML page\nor by making an API request for it.\n\n## Val Town Platform Specifics\n\n- **Redirects:** Use `return new Response(null, { status: 302, headers: { Location: \"/place/to/redirect\" }})` instead of `Response.redirect` which is broken\n- **Images:** Avoid external images or base64 images. Use emojis, unicode symbols, or icon fonts/libraries instead\n- **AI Image:** To inline generate an AI image use: `

`\n- **Storage:** DO NOT use the Deno KV module for storage\n- **Browser APIs:** DO NOT use the `alert()`, `prompt()`, or `confirm()` methods\n- **Weather Data:** Use open-meteo for weather data (doesn't require API keys) unless otherwise specified\n- **View Source:** Add a view source link by importing & using `import.meta.url.replace(\"ems.sh\", \"val.town)\"` (or passing this data to the client) and include `target=\"_top\"` attribute\n- **Error Debugging:** Add `` to HTML to capture client-side errors\n- **Error Handling:** Only use try...catch when there's a clear local resolution; Avoid catches that merely log or return 500s. Let errors bubble up with full context\n- **Environment Variables:** Use `Deno.env.get('keyname')` when you need to, but generally prefer APIs that don't require keys\n- **Imports:** Use `https://esm.sh` for npm and Deno dependencies to ensure compatibility on server and browser\n- **Storage Strategy:** Only use backend storage if explicitly required; prefer simple static client-side sites\n- **React Configuration:** When using React libraries, pin versions with `?deps=react@18.2.0,react-dom@18.2.0` and start the file with `/** @jsxImportSource https://esm.sh/react@18.2.0 */`\n- Ensure all React dependencies and sub-dependencies are pinned to the same version\n- **Styling:** Default to using TailwindCSS via `` unless otherwise specified\n\n## Project Structure and Design Patterns\n\n### Recommended Directory Structure\n```\n├── backend/\n│ ├── database/\n│ │ ├── migrations.ts # Schema definitions\n│ │ ├── queries.ts # DB query functions\n│ │ └── README.md\n│ └── routes/ # Route modules\n│ ├── [route].ts\n│ └── static.ts # Static file serving\n│ ├── index.ts # Main entry point\n│ └── README.md\n├── frontend/\n│ ├── components/\n│ │ ├── App.tsx\n│ │ └── [Component].tsx\n│ ├── favicon.svg\n│ ├── index.html # Main HTML template\n│ ├── index.tsx # Frontend JS entry point\n│ ├── README.md\n│ └── style.css\n├── README.md\n└── shared/\n ├── README.md\n └── utils.ts # Shared types and functions\n```\n\n### Backend (Hono) Best Practices\n\n- Hono is the recommended API framework\n- Main entry point should be `backend/index.ts`\n- **Static asset serving:** Use the utility functions to read and serve project files:\n ```ts\n import { readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n \n // serve all files in frontend/ and shared/\n app.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\n app.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n \n // For index.html, often you'll want to bootstrap with initial data\n app.get(\"/\", async c => {\n let html = await readFile(\"/frontend/index.html\", import.meta.url);\n \n // Inject data to avoid extra round-trips\n const initialData = await fetchInitialData();\n const dataScript = ``;\n \n html = html.replace(\"\", `${dataScript}`);\n return c.html(html);\n });\n ```\n- Create RESTful API routes for CRUD operations\n- Always include this snippet at the top-level Hono app to re-throwing errors to see full stack traces:\n ```ts\n // Unwrap Hono errors to see original error details\n app.onError((err, c) => {\n throw err;\n });\n ```\n\n### Database Patterns\n- Run migrations on startup or comment out for performance\n- Change table names when modifying schemas rather than altering\n- Export clear query functions with proper TypeScript typing\n\n## Common Gotchas and Solutions\n\n1. **Environment Limitations:** \n - Val Town runs on Deno in a serverless context, not Node.js\n - Code in `shared/` must work in both frontend and backend environments\n - Cannot use `Deno` keyword in shared code\n - Use `https://esm.sh` for imports that work in both environments\n\n2. **SQLite Peculiarities:**\n - Limited support for ALTER TABLE operations\n - Create new tables with updated schemas and copy data when needed\n - Always run table creation before querying\n\n3. **React Configuration:**\n - All React dependencies must be pinned to 18.2.0\n - Always include `@jsxImportSource https://esm.sh/react@18.2.0` at the top of React files\n - Rendering issues often come from mismatched React versions\n\n4. **File Handling:**\n - Val Town only supports text files, not binary\n - Use the provided utilities to read files across branches and forks\n - For files in the project, use `readFile` helpers\n\n5. **API Design:**\n - `fetch` handler is the entry point for HTTP vals\n - Run the Hono app with `export default app.fetch // This is the entry point for HTTP vals`\n\n",[],[397],{"_106":405,"_97":406,"_165":407,"_167":-5,"_168":150,"_169":408,"_171":121,"_108":409,"_102":300,"_173":205,"_175":410,"_176":22,"_177":-5,"_178":411,"_100":412},"8258aa2a-f013-11f0-8b0c-42dde27851f2","App.tsx","c9ecae0c-d7b0-11f0-ae93-42dde27851f2",1093,["D",1768262703031],"/** @jsxImportSource https://esm.sh/react */\nimport React, { useEffect, useState } from \"https://esm.sh/react\";\nimport type {\n MasterPasswordMode,\n Screen,\n} from \"./types.ts\";\nimport { useWebLLM } from \"./hooks/useWebLLM.ts\";\nimport { useEncryption } from \"./hooks/useEncryption.ts\";\nimport {\n createMasterPasswordToken,\n hasMasterPassword,\n verifyMasterPassword,\n} from \"./lib/crypto.ts\";\nimport { MainScreen } from \"./screens/MainScreen.tsx\";\nimport { MasterPasswordScreen } from \"./screens/MasterPasswordScreen.tsx\";\nimport { UnsupportedScreen } from \"./screens/UnsupportedScreen.tsx\";\nimport {\n isMobileDevice,\n isWebGPUFunctional,\n} from \"./lib/deviceDetection.ts\";\n\nexport function App() {\n const [screen, setScreen] = useState(\"masterpassword\");\n const [masterPasswordMode, setMasterPasswordMode] = useState<\n MasterPasswordMode\n >(\"create\");\n const [error, setError] = useState(null);\n const [isLoading, setIsLoading] = useState(false);\n const [isCheckingPassword, setIsCheckingPassword] = useState(true);\n const [blockingReason, setBlockingReason] = useState<\n \"mobile\" | \"no-webgpu\" | null\n >(null);\n const [selectedModel, setSelectedModel] = useState(() => {\n // Load saved model from localStorage or use default\n const savedModel = localStorage.getItem(\"selectedModel\");\n return savedModel || \"Llama-3.2-3B-Instruct-q4f16_1-MLC\"; // Medium default\n });\n\n const { status, isReady, sendMessage } = useWebLLM(selectedModel);\n const { unlock, lock, encryptMessages, decryptMessages } = useEncryption();\n\n // Save selected model to localStorage when it changes\n useEffect(() => {\n localStorage.setItem(\"selectedModel\", selectedModel);\n }, [selectedModel]);\n\n // Check for unsupported environment (mobile or no WebGPU)\n useEffect(() => {\n async function checkEnvironment() {\n // Check mobile first (synchronous)\n if (isMobileDevice()) {\n setBlockingReason(\"mobile\");\n setScreen(\"unsupported\");\n setIsCheckingPassword(false);\n return;\n }\n\n // Then test if WebGPU is actually functional (async)\n const webGPUWorks = await isWebGPUFunctional();\n if (!webGPUWorks) {\n setBlockingReason(\"no-webgpu\");\n setScreen(\"unsupported\");\n setIsCheckingPassword(false);\n }\n }\n\n checkEnvironment();\n }, []);\n\n // Check password status on mount\n useEffect(() => {\n async function checkPassword() {\n try {\n const hasPassword = await hasMasterPassword();\n setMasterPasswordMode(hasPassword ? \"verify\" : \"create\");\n\n // Check if already unlocked in session\n const storedPassword = sessionStorage.getItem(\"masterPassword\");\n if (storedPassword && hasPassword) {\n const isValid = await verifyMasterPassword(storedPassword);\n if (isValid) {\n unlock(storedPassword);\n setScreen(\"main\");\n }\n }\n } catch (error) {\n console.error(\"Failed to check password status:\", error);\n setMasterPasswordMode(\"create\");\n } finally {\n setIsCheckingPassword(false);\n }\n }\n\n checkPassword();\n }, [unlock]);\n\n // Handle master password creation (first time)\n const handleCreateMasterPassword = async (password: string) => {\n setError(null);\n setIsLoading(true);\n\n try {\n await createMasterPasswordToken(password);\n unlock(password);\n sessionStorage.setItem(\"masterPassword\", password);\n setScreen(\"main\");\n } catch (e: any) {\n setError(e.message);\n }\n\n setIsLoading(false);\n };\n\n // Handle master password verification (returning user)\n const handleVerifyMasterPassword = async (password: string) => {\n setError(null);\n setIsLoading(true);\n\n try {\n const isValid = await verifyMasterPassword(password);\n if (!isValid) {\n setError(\"Wrong password\");\n setIsLoading(false);\n return;\n }\n\n unlock(password);\n sessionStorage.setItem(\"masterPassword\", password);\n setScreen(\"main\");\n } catch (e: any) {\n setError(e.message);\n }\n\n setIsLoading(false);\n };\n\n // Handle lock app\n const handleLock = () => {\n lock();\n sessionStorage.removeItem(\"masterPassword\");\n setScreen(\"masterpassword\");\n setMasterPasswordMode(\"verify\");\n setError(null);\n };\n\n // Show loading screen while checking password status\n if (isCheckingPassword) {\n return (\n

\n );\n }\n\n // Render unsupported screen for mobile or browsers without WebGPU\n if (screen === \"unsupported\") {\n return ;\n }\n\n // Render master password screen\n if (screen === \"masterpassword\") {\n return (\n \n );\n }\n\n // Render main screen\n return (\n \n );\n}\n",[],[406],{"_106":414,"_97":415,"_165":416,"_167":-5,"_168":150,"_169":417,"_171":121,"_108":418,"_102":235,"_173":391,"_175":419,"_176":22,"_177":-5,"_178":420,"_100":421},"493b04f4-ea6e-11f0-8472-42dde27851f2","README.md","0c7422e8-dac5-11f0-8589-42dde27851f2",884,["D",1767641984422],"# Locals AI 🤖\n\nA local, private AI chat running entirely on your browser and saved on your Val\nTown. Your conversations are encrypted client-side before being stored, and your\npassword never leaves your device.\n\n## Features\n\n- 🔒 **Client-side encryption** - All chats encrypted with AES-256-GCM before\n storage\n- 🔑 **Master password protection** - Single password protects all your\n conversations\n- 💾 **Self-hosted on Val Town** - Your data stays in your own Val Town SQLite\n database\n- 🚫 **No tracking** - No analytics, no third parties\n- 🧠 **Local AI** - Uses WebLLM to run AI models directly in your browser\n- 🎛️ **Multiple AI models** - Choose between light, medium, and large models\n- 🍴 **Fork-friendly** - Easy to remix and run your own instance\n\n## How It Works\n\n1. **Set a master password** on first use to protect all your chats\n2. **Create unlimited chats** with custom names\n3. **Your messages are encrypted** in the browser using your master password\n4. **Encrypted data is stored** in your Val Town SQLite database\n5. **Only you can decrypt** your chats with the correct password\n\nYour password is never sent to the server. Encryption happens entirely in your\nbrowser.\n\n## Quick Start: Remix This Val\n\n### Step 1: Create a Val.Town Account\n\nIf you don't already have one, create a free account on [Val.Town](https://val.town).\n\n### Step 2: Fork the Val\n\n1. Click the \"Remix\" button on Val Town (or fork this repository)\n2. This creates your own copy of the entire project with all files (`main.ts`, `api/storage.ts`, etc.)\n\n**Note:** In Val.Town, this is **one val/project** with multiple files. Each file that exports a default function gets its own HTTP endpoint when Val.Town serves it.\n\n### Step 3: Find Your Storage API Endpoint\n\nAfter forking, you need to find the HTTP endpoint for your `api/storage.ts` file:\n\n1. In your forked val, navigate to the `api/storage.ts` file\n2. Val.Town will show you the **HTTP endpoint URL** for this file (usually displayed at the top)\n3. **Copy this URL** - it will look something like:\n - `https://yourname--a1b2c3d4e5f6.web.val.run`\n\n (The actual format depends on how Val.Town generates endpoints for your val files)\n\n4. This is your `STORAGE_URL` - you'll need it in the next step\n\n### Step 4: Set Up Environment Variables\n\nNow configure the environment variables for your entire val/project.\n\n**Important:** Environment variables in Val.Town are **global for the entire project**, not per-file. When you set them, they apply to both `main.ts` and `api/storage.ts`.\n\n#### Steps:\n\n1. Visit your main val's URL (the `main.ts` endpoint) in your browser\n2. Check the **browser console** or the **Val.Town logs**. You'll see messages like:\n\n ```\n ⚠️ STORAGE_API_KEY not set! Auto-generated key: a1b2c3d4-e5f6-7890-abcd-ef1234567890\n ⚠️ Save this key to your STORAGE_API_KEY environment variable!\n ❌ STORAGE_URL not set! Please set it in your environment variables.\n ```\n\n3. In your Val.Town project settings, go to **Environment Variables**\n4. Add these two variables:\n - **Name:** `STORAGE_API_KEY`\n **Value:** The auto-generated key from the console logs\n - **Name:** `STORAGE_URL`\n **Value:** The HTTP endpoint URL for `api/storage.ts` (from Step 3)\n Example: `https://yourname--a1b2c3d4e5f6.web.val.run`\n\n5. Save the environment variables\n\nSince environment variables are global, both `main.ts` and `api/storage.ts` will now have access to the same `STORAGE_API_KEY` and can communicate securely.\n\n### Step 5: Use Your App\n\nVisit your main val's URL (the `main.ts` endpoint):\n\n1. Set a strong master password on first visit\n2. Create your first chat with the \"+ New Chat\" button\n3. Select your preferred AI model from the dropdown\n4. Start chatting - all messages are automatically encrypted and saved!\n\n## Security Features\n\n### What's Protected\n\n✅ **SQL Injection Prevention** - All database queries use parameterized\nstatements ✅ **API Authentication** - Storage endpoint requires API key ✅\n**Client-side Encryption** - AES-256-GCM with PBKDF2 key derivation ✅ **Master\nPassword Protection** - Single password encrypts all conversations ✅ **Session\nManagement** - Password stored only in sessionStorage, cleared on lock\n\n### Encryption Details\n\n- **Algorithm:** AES-GCM (256-bit)\n- **Key Derivation:** PBKDF2 with 100,000 iterations\n- **Random Salt & IV:** Generated for each encryption\n- **Password:** Never leaves your device\n\n### What's Stored in the Database\n\nYour Val Town SQLite database stores:\n\n- Chat ID (random UUID)\n- Chat name (plain text)\n- Encrypted chat data (ciphertext + salt + IV)\n- Timestamps\n\n**Without your password, the encrypted data is unreadable.**\n\n## Environment Variables\n\n| Variable | Required | Description |\n| ----------------- | -------- | -------------------------------------------------------------------------------- |\n| `STORAGE_URL` | **Yes** | URL of your storage val (e.g., `https://yourname--storage.web.val.run`) |\n| `STORAGE_API_KEY` | **Yes** | Protects your storage API from unauthorized access (auto-generated on first run) |\n| `FRONTEND_URL` | No | Optional CORS restriction (e.g., `https://yourapp.val.run`) |\n\n## Architecture\n\n```\n┌─────────────────┐\n│ Browser │\n│ (React App) │\n│ │\n│ • WebLLM AI │\n│ • Encryption │\n│ • UI │\n└────────┬────────┘\n │\n │ HTTPS + API Key\n │\n┌────────▼────────┐\n│ Val Town │\n│ Storage API │\n│ │\n│ • SQLite DB │\n│ • Auth Check │\n└─────────────────┘\n```\n\n## File Structure\n\n```\n├── main.ts # Entry point, serves HTML\n├── App.tsx # Main app with master password flow\n├── api/\n│ └── storage.ts # Storage API with authentication\n├── lib/\n│ ├── api.ts # Client-side API calls\n│ ├── crypto.ts # Encryption/decryption logic\n│ └── models.ts # AI model definitions\n├── hooks/\n│ ├── useEncryption.ts # Encryption state management\n│ └── useWebLLM.ts # AI model integration\n├── screens/\n│ ├── MasterPasswordScreen.tsx # Password creation/verification\n│ └── MainScreen.tsx # Main chat interface with sidebar\n└── components/\n ├── MessageBubble.tsx # Message display\n ├── StatusBadge.tsx # AI status indicator\n ├── EmptyState.tsx # Empty state UI\n ├── ModelSelector.tsx # AI model selection dropdown\n ├── OptionsMenu.tsx # Lock/delete options menu\n └── ChatNameModal.tsx # Create/rename chat modal\n```\n\n## Usage Tips\n\n### Strong Master Password\n\nUse a strong, unique master password. Since encryption happens client-side, your\nmaster password is the only thing protecting ALL your data.\n\n### Password Recovery\n\n**There is no password recovery.** If you forget your master password, ALL your\nchat data is permanently inaccessible. This is by design - the server (in this\ncase Val-Town) don't have access to your data or password.\n\n### AI Model Selection\n\nThe app offers three AI models:\n\n- **Phi-3 Mini (2.3GB)** - Fastest, best for quick responses\n- **Llama 3.2 3B (1.9GB)** - Balanced performance and quality (default)\n- **Llama 3.1 8B (4.7GB)** - Highest quality responses, slower\n\nModels download to your browser once and persist. You can switch models anytime\nin the chat interface.\n\n## Troubleshooting\n\n### \"Unauthorized\" Error\n\n**Problem:** API calls return 401 Unauthorized\n\n**Solution:**\n\n1. Check that `STORAGE_API_KEY` is set in environment variables\n2. Make sure the key matches in both your main val and storage val\n3. Restart both vals after setting the variable\n\n### \"Wrong Password\" Error\n\n**Problem:** Can't unlock the app\n\n**Solution:**\n\n- Double-check your master password (passwords are case-sensitive)\n- There is no password recovery - if you forgot it, all your chats are\n inaccessible\n\n### Chat List Empty\n\n**Problem:** No chats appear on home screen\n\n**Solution:**\n\n1. Check browser console for errors\n2. Verify `STORAGE_URL` environment variable is set correctly in Val Town\n settings\n3. Ensure storage val is running and accessible\n4. Check that both vals have the same `STORAGE_API_KEY`\n\n### AI Model Not Loading\n\n**Problem:** WebLLM status shows error\n\n**Solution:**\n\n- WebLLM downloads models to browser (can be 1-2 GB)\n- Check your internet connection\n- Try refreshing the page\n- Check browser console for specific errors\n\n## Advanced Configuration\n\n### Change PBKDF2 Iterations\n\nFor stronger encryption, increase iterations in `lib/crypto.ts` (line 25):\n\n```typescript\niterations: 600000, // Increased from 100,000\n```\n\n**Trade-off:** Higher security, slower encryption/decryption.\n\n### Restrict CORS\n\nTo only allow requests from your frontend:\n\n1. Set `FRONTEND_URL` environment variable\n2. Value: `https://your-main-val.val.run`\n3. Restart storage val\n\nThis prevents other websites from calling your API (browser-level protection).\n\n### Add More AI Models\n\nTo add additional AI models beyond the three included:\n\n1. Edit `lib/models.ts` to add model definitions\n2. Add to the `AVAILABLE_MODELS` array with model ID, name, size, and\n description\n3. Find compatible model IDs in the\n [WebLLM documentation](https://github.com/mlc-ai/web-llm)\n\n### Lock and Security Features\n\n- **Lock App**: Click the options menu (•••) in the bottom-right to lock the\n app. This clears your password from the session and returns to the master\n password screen.\n- **Delete All Data**: Use the options menu to permanently delete all chats from\n your database. This action cannot be undone.\n\n## Privacy & Data\n\n### What Data is Collected\n\n**None.** This app doesn't collect, track, or transmit any data beyond what's\nnecessary for functionality:\n\n- Your chats are encrypted and stored in **your** Val Town SQLite database\n- Your password never leaves your browser\n- No analytics, no tracking, no third parties\n\n### Who Can Access Your Data\n\n**Only you.** When you fork this val:\n\n- You own the Val Town vals\n- You control the database\n- You set the API key\n- You hold the encryption passwords\n\nNo one else (including the original author) can access your chats.\n\n## Contributing\n\nFound a bug or want to improve the code? Feel free to:\n\n1. Fork this val\n2. Make your improvements\n3. Share your enhanced version!\n\nThis is open source - make it your own.\n\n## Security Considerations\n\nWhile this app implements strong encryption and security practices, please note:\n\n- ⚠️ This is a personal project, not audited by security professionals\n- ⚠️ API keys are visible in browser DevTools (use strong, unique keys)\n- ⚠️ Val Town is a public platform - keep your vals private if needed\n- ⚠️ No rate limiting implemented (could be added if needed)\n\nFor maximum security:\n\n- Use strong, unique passwords\n- Set a strong `STORAGE_API_KEY`\n- Keep your Val Town account secure\n- Consider enabling 2FA on your Val Town account\n\n## Credits\n\nBuilt with:\n\n- [Val Town](https://val.town) - Hosting & SQLite\n- [React](https://react.dev) - UI framework\n- [WebLLM](https://github.com/mlc-ai/web-llm) - Browser-based AI\n- [Web Crypto API](https://developer.mozilla.org/en-US/docs/Web/API/Web_Crypto_API) -\n Encryption\n- [Tailwind CSS](https://tailwindcss.com) - Styling\n\n## License\n\nMIT License - Feel free to fork, remix, and make it your own!\n\n---\n\n**Remember:** Your master password is your key. Keep it safe, keep it secret.\n",[],[415],{"_106":423,"_97":424,"_165":425,"_167":-5,"_168":150,"_169":326,"_171":121,"_108":426,"_102":33,"_173":391,"_175":427,"_176":22,"_177":-5,"_178":428,"_100":429},"1ea7e45c-d7ac-11f0-b406-42dde27851f2","deno.json","1ea7e52e-d7ac-11f0-b406-42dde27851f2",["D",1765579469700],"{\n \"$schema\": \"https://raw.githubusercontent.com/denoland/deno/98f62cee78e85bfc47c62ed703777c6bc8794f1c/cli/schemas/config-file.v1.json\",\n \"lock\": false,\n \"compilerOptions\": {\n \"noImplicitAny\": false,\n \"strict\": true,\n \"types\": [\"https://www.val.town/types/valtown.d.ts\"],\n \"lib\": [\n \"dom\",\n \"dom.iterable\",\n \"dom.asynciterable\",\n \"deno.ns\",\n \"deno.unstable\"\n ]\n },\n \"lint\": {\n \"include\": [\"deno:/https/esm.town/**/*\"],\n \"rules\": {\n \"exclude\": [\"no-explicit-any\"]\n }\n },\n \"node_modules_dir\": false,\n \"experimental\": {\n \"unstable-node-globals\": true,\n \"unstable-temporal\": true,\n \"unstable-worker-options\": true,\n \"unstable-sloppy-imports\": true\n }\n}",[],[424],{"_106":431,"_97":432,"_165":119,"_167":-5,"_168":150,"_169":433,"_171":22,"_108":434,"_102":435,"_173":187,"_175":436,"_176":22,"_177":-5,"_178":437,"_100":438},"18ba0c16-f014-11f0-a6a7-42dde27851f2","main.ts",1098,["D",1768262955321],31,"/** @jsxImportSource https://esm.sh/react */\nimport React from \"https://esm.sh/react\";\nimport { createRoot } from \"https://esm.sh/react-dom/client\";\nimport { App } from \"./App.tsx\";\n\nif (typeof document !== \"undefined\") {\n console.log(\"[main] Hydrating...\");\n createRoot(document.getElementById(\"root\")).render();\n}\n\nexport default async function(req: Request): Promise {\n // Get or auto-generate API key (same logic as storage API)\n let storageApiKey = Deno.env.get(\"STORAGE_API_KEY\");\n if (!storageApiKey) {\n storageApiKey = crypto.randomUUID();\n console.warn(\"⚠️ STORAGE_API_KEY not set in main.ts! Auto-generated key:\", storageApiKey);\n console.warn(\"⚠️ Save this key to your STORAGE_API_KEY environment variable!\");\n }\n\n // Get storage URL from environment\n const storageUrl = Deno.env.get(\"STORAGE_URL\");\n if (!storageUrl) {\n console.error(\"❌ STORAGE_URL not set! Please set it in your environment variables.\");\n console.error(\" Example: https://your-username--your-storage-val.web.val.run\");\n }\n\n const html = `\n\n\n \n \n Private AI Chat\n \n \n \n \n \n \n \n\n\n

\n \n\n`;\n\n return new Response(html, {\n headers: { \"content-type\": \"text/html\" }\n });\n}",[],[432],{"_106":440,"_97":441,"_165":442,"_167":-5,"_168":150,"_169":443,"_171":121,"_108":444,"_102":399,"_173":205,"_175":445,"_176":22,"_177":-5,"_178":446,"_100":447},"b74d6686-dd48-11f0-a31c-42dde27851f2","tailwind.config.js","e9c2c03c-dd36-11f0-bddc-42dde27851f2",709,["D",1766196483106],"/** @type {import('tailwindcss').Config} */\nmodule.exports = {\n content: [\n \"./**/*.{ts,tsx,js,jsx}\",\n \"./screens/**/*.{ts,tsx}\",\n \"./components/**/*.{ts,tsx}\",\n ],\n theme: {\n extend: {\n fontFamily: {\n sans: [\n \"Inter\",\n \"-apple-system\",\n \"BlinkMacSystemFont\",\n \"Segoe UI\",\n \"Roboto\",\n \"Helvetica Neue\",\n \"Arial\",\n \"sans-serif\",\n ],\n logo: [\"Baskervville\", \"Georgia\", \"serif\"],\n reading: [\"Lora\", \"Georgia\", \"Times New Roman\", \"serif\"],\n },\n colors: {\n primary: {\n 100: \"#79C5FC\",\n 200: \"#42B0FF\",\n 300: \"#3E95CF\",\n 400: \"#1D618F\",\n },\n vibe: {\n 100: \"#CCCCCC\",\n 200: \"#999999\",\n 250: \"#5D5C56\",\n 300: \"#30302E\",\n 400: \"#262624\",\n 500: \"#202124\",\n 600: \"#18191B\",\n 700: \"#141413\",\n },\n },\n },\n },\n plugins: [],\n};\n",[],[441],{"_106":449,"_97":450,"_165":451,"_167":-5,"_168":150,"_169":452,"_171":22,"_108":453,"_102":352,"_173":205,"_175":454,"_176":22,"_177":-5,"_178":455,"_100":456},"ac3a7e9e-008b-11f1-b4ef-42dde27851f2","types.ts","c9fc96be-d7b0-11f0-ae93-42dde27851f2",1147,["D",1770073531505],"export type Message = {\n role: \"user\" | \"assistant\";\n content: string;\n thinking?: string;\n};\n\nexport type Screen = \"masterpassword\" | \"main\" | \"unsupported\";\n\nexport type MasterPasswordMode = \"create\" | \"verify\";\n\nexport type MasterPasswordToken = {\n version: number;\n verificationData: EncryptedPackage;\n createdAt: string;\n};\n\nexport type WebLLMStatus = {\n state: \"idle\" | \"loading\" | \"ready\" | \"error\";\n message: string;\n progress?: number;\n};\n\nexport type EncryptedPackage = {\n salt: number[];\n iv: number[];\n ciphertext: number[];\n};\n\nexport type ChatListItem = {\n id: string;\n name: string;\n createdAt: string;\n updatedAt: string;\n};\n\n/**\n * WebLLM engine instance interface\n */\nexport interface WebLLMEngine {\n chat: {\n completions: {\n create(request: WebLLMCompletionRequest): Promise>;\n };\n };\n unload(): Promise;\n}\n\n/**\n * Progress information during model loading\n */\nexport interface WebLLMProgress {\n progress: number; // 0.0 to 1.0\n text?: string;\n timeElapsed?: number;\n}\n\n/**\n * Callback function for tracking model loading progress\n */\nexport type WebLLMProgressCallback = (progress: WebLLMProgress) => void;\n\n/**\n * Request parameters for WebLLM chat completion\n */\nexport interface WebLLMCompletionRequest {\n messages: Array<{\n role: 'system' | 'user' | 'assistant';\n content: string;\n }>;\n temperature?: number;\n max_tokens?: number;\n stream?: boolean;\n}\n\n/**\n * Response from WebLLM chat completion\n */\nexport interface WebLLMCompletionResponse {\n choices: Array<{\n message: {\n role: 'assistant';\n content: string;\n };\n finish_reason: string;\n }>;\n}\n\n/**\n * Internal configuration for WebLLM hook\n */\nexport interface WebLLMConfig {\n cdnUrl: string;\n systemPrompt: string;\n temperature: number;\n maxTokens: number;\n retryDelayMs: number;\n maxRetryAttempts: number;\n}\n\n/**\n * Callback for streaming updates during message generation\n */\nexport type StreamCallback = (update: {\n phase: \"thinking\" | \"response\";\n content: string;\n isComplete: boolean;\n}) => void;\n\n/**\n * Streaming chunk from WebLLM\n */\nexport interface WebLLMStreamChunk {\n choices: Array<{\n delta: { content?: string; role?: string };\n finish_reason?: string | null;\n }>;\n}\n\n/**\n * Options for sendMessage with streaming support\n */\nexport interface SendMessageOptions {\n enableThinking?: boolean;\n onStream?: StreamCallback;\n}\n\n/**\n * Result from sendMessage with optional thinking content\n */\nexport interface SendMessageResult {\n content: string;\n thinking?: string;\n}",[],[450],"files",[162,180,192,199,209,219,229,239,249,259,269,275,285,296,303,312,320,330,339,349,355,364,374,385,395,404,413,422,430,439,448],"counts",{"_461":33,"_462":196,"_463":33,"_464":465,"_468":469},"pulls","branches","forks","envVars",{"_466":399,"_467":33},"count","emptyCount","connections",{"_466":33,"_470":33},"unfilledCount","contents",[473,476,479],{"_106":356,"_97":357,"_165":358,"_167":351,"_168":150,"_169":103,"_171":22,"_108":474,"_102":360,"_173":205,"_175":361,"_176":22,"_177":-5,"_100":475},["D",1770073650143],[101,357],{"_106":365,"_97":366,"_165":367,"_167":351,"_168":150,"_169":368,"_171":121,"_108":477,"_102":370,"_173":205,"_175":371,"_176":22,"_177":-5,"_100":478},["D",1768262790810],[101,366],{"_106":375,"_97":376,"_165":377,"_167":351,"_168":150,"_169":378,"_171":121,"_108":480,"_102":380,"_173":205,"_175":381,"_176":22,"_177":-5,"_100":481},["D",1768003111860],[101,376],{"_106":350,"_97":101,"_165":351,"_167":-5,"_168":150,"_169":352,"_171":121,"_108":483,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_100":484},["D",1765581474705],[101],"upstreamProject","upstreamBranch","emptyEnvVars","isNew","flags",[],"routes/_app.x.$handle.$name.branch.$branch.version.$version.code.$",{"_95":96,"_97":98,"_99":45,"_102":103,"_100":101,"_104":493,"_160":498,"_457":592,"_391":593,"_471":595,"_602":-5,"_603":-5,"_604":-5,"_605":-5,"_606":607,"_464":-5},{"_106":107,"_97":98,"_108":494,"_110":-5,"_111":22,"_112":113,"_114":115,"_116":-5,"_14":117,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":33,"_16":125,"_126":127,"_128":129,"_130":131,"_132":131,"_133":134,"_135":134,"_136":134,"_137":131,"_138":131,"_23":495,"_142":496,"_147":497,"_159":22},["D",1765579434547],{"_95":96,"_140":141},{"_95":96,"_144":141,"_14":117,"_145":146},{"_102":149,"_97":45,"_106":150,"_151":107,"_152":-5,"_153":-5,"_154":-5,"_155":-5,"_156":-5,"_157":98,"_158":96},[499,505,529,538,556,568,571,574,577,580,583,586,589],{"_106":163,"_97":164,"_165":166,"_167":-5,"_168":150,"_169":170,"_171":121,"_108":172,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":500,"_100":504},[501],{"_106":181,"_97":182,"_165":183,"_167":166,"_168":150,"_169":184,"_171":121,"_108":185,"_102":186,"_173":187,"_175":188,"_176":22,"_177":-5,"_178":502,"_100":503},[],[164,182],[164],{"_106":193,"_97":194,"_165":195,"_167":-5,"_168":150,"_169":196,"_171":121,"_108":197,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":506,"_100":528},[507,510,513,516,519,522,525],{"_106":200,"_97":201,"_165":202,"_167":195,"_168":150,"_169":203,"_171":121,"_108":204,"_102":196,"_173":205,"_175":206,"_176":22,"_177":-5,"_178":508,"_100":509},[],[194,201],{"_106":210,"_97":211,"_165":212,"_167":195,"_168":150,"_169":213,"_171":121,"_108":214,"_102":215,"_173":205,"_175":216,"_176":22,"_177":-5,"_178":511,"_100":512},[],[194,211],{"_106":220,"_97":221,"_165":222,"_167":195,"_168":150,"_169":223,"_171":22,"_108":224,"_102":225,"_173":205,"_175":226,"_176":22,"_177":-5,"_178":514,"_100":515},[],[194,221],{"_106":230,"_97":231,"_165":232,"_167":195,"_168":150,"_169":233,"_171":121,"_108":234,"_102":235,"_173":205,"_175":236,"_176":22,"_177":-5,"_178":517,"_100":518},[],[194,231],{"_106":240,"_97":241,"_165":242,"_167":195,"_168":150,"_169":243,"_171":121,"_108":244,"_102":245,"_173":205,"_175":246,"_176":22,"_177":-5,"_178":520,"_100":521},[],[194,241],{"_106":250,"_97":251,"_165":252,"_167":195,"_168":150,"_169":253,"_171":121,"_108":254,"_102":255,"_173":205,"_175":256,"_176":22,"_177":-5,"_178":523,"_100":524},[],[194,251],{"_106":260,"_97":261,"_165":262,"_167":195,"_168":150,"_169":263,"_171":22,"_108":264,"_102":33,"_173":205,"_175":265,"_176":22,"_177":-5,"_178":526,"_100":527},[],[194,261],[194],{"_106":270,"_97":271,"_165":272,"_167":-5,"_168":150,"_169":225,"_171":121,"_108":273,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":530,"_100":537},[531,534],{"_106":276,"_97":277,"_165":278,"_167":272,"_168":150,"_169":279,"_171":121,"_108":280,"_102":281,"_173":205,"_175":282,"_176":22,"_177":-5,"_178":532,"_100":533},[],[271,277],{"_106":286,"_97":287,"_165":288,"_167":272,"_168":150,"_169":289,"_171":22,"_108":290,"_102":291,"_173":205,"_175":292,"_176":22,"_177":-5,"_178":535,"_100":536},[],[271,287],[271],{"_106":297,"_97":298,"_165":299,"_167":-5,"_168":150,"_169":300,"_171":121,"_108":301,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":539,"_100":555},[540,543,546,549,552],{"_106":304,"_97":305,"_165":306,"_167":299,"_168":150,"_169":184,"_171":121,"_108":307,"_102":308,"_173":205,"_175":309,"_176":22,"_177":-5,"_178":541,"_100":542},[],[298,305],{"_106":313,"_97":314,"_165":315,"_167":299,"_168":150,"_169":184,"_171":121,"_108":316,"_102":281,"_173":205,"_175":317,"_176":22,"_177":-5,"_178":544,"_100":545},[],[298,314],{"_106":321,"_97":322,"_165":323,"_167":299,"_168":150,"_169":324,"_171":121,"_108":325,"_102":326,"_173":205,"_175":327,"_176":22,"_177":-5,"_178":547,"_100":548},[],[298,322],{"_106":331,"_97":332,"_165":333,"_167":299,"_168":150,"_169":334,"_171":121,"_108":335,"_102":33,"_173":205,"_175":336,"_176":22,"_177":-5,"_178":550,"_100":551},[],[298,332],{"_106":340,"_97":341,"_165":342,"_167":299,"_168":150,"_169":343,"_171":121,"_108":344,"_102":33,"_173":205,"_175":345,"_176":22,"_177":-5,"_178":553,"_100":554},[],[298,341],[298],{"_106":350,"_97":101,"_165":351,"_167":-5,"_168":150,"_169":352,"_171":121,"_108":353,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_178":557,"_100":567},[558,561,564],{"_106":356,"_97":357,"_165":358,"_167":351,"_168":150,"_169":103,"_171":22,"_108":359,"_102":360,"_173":205,"_175":361,"_176":22,"_177":-5,"_178":559,"_100":560},[],[101,357],{"_106":365,"_97":366,"_165":367,"_167":351,"_168":150,"_169":368,"_171":121,"_108":369,"_102":370,"_173":205,"_175":371,"_176":22,"_177":-5,"_178":562,"_100":563},[],[101,366],{"_106":375,"_97":376,"_165":377,"_167":351,"_168":150,"_169":378,"_171":121,"_108":379,"_102":380,"_173":205,"_175":381,"_176":22,"_177":-5,"_178":565,"_100":566},[],[101,376],[101],{"_106":386,"_97":387,"_165":388,"_167":-5,"_168":150,"_169":389,"_171":121,"_108":390,"_102":281,"_173":391,"_175":392,"_176":22,"_177":-5,"_178":569,"_100":570},[],[387],{"_106":396,"_97":397,"_165":398,"_167":-5,"_168":150,"_169":399,"_171":121,"_108":400,"_102":33,"_173":391,"_175":401,"_176":22,"_177":-5,"_178":572,"_100":573},[],[397],{"_106":405,"_97":406,"_165":407,"_167":-5,"_168":150,"_169":408,"_171":121,"_108":409,"_102":300,"_173":205,"_175":410,"_176":22,"_177":-5,"_178":575,"_100":576},[],[406],{"_106":414,"_97":415,"_165":416,"_167":-5,"_168":150,"_169":417,"_171":121,"_108":418,"_102":235,"_173":391,"_175":419,"_176":22,"_177":-5,"_178":578,"_100":579},[],[415],{"_106":423,"_97":424,"_165":425,"_167":-5,"_168":150,"_169":326,"_171":121,"_108":426,"_102":33,"_173":391,"_175":427,"_176":22,"_177":-5,"_178":581,"_100":582},[],[424],{"_106":431,"_97":432,"_165":119,"_167":-5,"_168":150,"_169":433,"_171":22,"_108":434,"_102":435,"_173":187,"_175":436,"_176":22,"_177":-5,"_178":584,"_100":585},[],[432],{"_106":440,"_97":441,"_165":442,"_167":-5,"_168":150,"_169":443,"_171":121,"_108":444,"_102":399,"_173":205,"_175":445,"_176":22,"_177":-5,"_178":587,"_100":588},[],[441],{"_106":449,"_97":450,"_165":451,"_167":-5,"_168":150,"_169":452,"_171":22,"_108":453,"_102":352,"_173":205,"_175":454,"_176":22,"_177":-5,"_178":590,"_100":591},[],[450],[499,501,505,507,510,513,516,519,522,525,529,531,534,538,540,543,546,549,552,556,558,561,564,568,571,574,577,580,583,586,589],{"_106":350,"_97":101,"_165":351,"_167":-5,"_168":150,"_169":352,"_171":121,"_108":483,"_102":33,"_173":174,"_175":-5,"_176":22,"_177":-5,"_100":594},[101],[596,598,600],{"_106":356,"_97":357,"_165":358,"_167":351,"_168":150,"_169":103,"_171":22,"_108":474,"_102":360,"_173":205,"_175":361,"_176":22,"_177":-5,"_100":597},[101,357],{"_106":365,"_97":366,"_165":367,"_167":351,"_168":150,"_169":368,"_171":121,"_108":477,"_102":370,"_173":205,"_175":371,"_176":22,"_177":-5,"_100":599},[101,366],{"_106":375,"_97":376,"_165":377,"_167":351,"_168":150,"_169":378,"_171":121,"_108":480,"_102":380,"_173":205,"_175":381,"_176":22,"_177":-5,"_100":601},[101,376],"upstream","pull","ahead","behind","templates",[],"actionData","errors"]

Prada

Locals-AI