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: `

Hi

`\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","2026-02-11T00:30:01.736Z","index-now","https://www.val.town/x/ashryanio/index-now/code/",[171,173,175,177,179,181,183,185,187,189],{"_129":172,"_131":132,"_133":22},122,{"_129":174,"_131":74,"_133":22},123,{"_129":176,"_131":138,"_133":139},124,{"_129":178,"_131":74,"_133":22},125,{"_129":180,"_131":144,"_133":22},126,{"_129":182,"_131":147,"_133":139},127,{"_129":184,"_131":150,"_133":139},128,{"_129":186,"_131":153,"_133":139},129,{"_129":188,"_131":156,"_133":22},130,{"_129":190,"_131":159,"_133":22},131,{"_106":192,"_108":193,"_110":194,"_112":195,"_114":196,"_116":197,"_118":198,"_120":199,"_122":200,"_124":201,"_125":202,"_126":203,"_160":273},"019a544d-22c4-74ad-9ffa-d14f4b401148","https://www.val.town/x/RobooHood/Free-Omni-Telegram/code/main.tsx","18a3ab06-ba4f-11f0-9425-0224a6c84d84","main.tsx","import {\n Bot,\n InputFile,\n webhookCallback,\n} from \"https://deno.land/x/grammy@v1.35.0/mod.ts\";\n\nif (!Deno.env.get(\"TELEGRAM_TOKEN\")) {\n throw new Error(\"TELEGRAM_TOKEN is not set\");\n}\nconst telegramToken = Deno.env.get(\"TELEGRAM_TOKEN\")!;\n\n// NVIDIA API as primary for text\nconst nvidiaApiKey = Deno.env.get(\"NVIDIA_API_KEY\");\n\n// OpenRouter as fallback\nconst openrouterApiKey = Deno.env.get(\"OPENROUTER_API_KEY\");\nconst openrouterApiKey2 = Deno.env.get(\"OPENROUTER_API_KEY2\");\n\n// Poe API (new OpenAI-compatible format)\nconst poeApiKey = Deno.env.get(\"POE_API_KEY\");\n\nconst bot = new Bot(telegramToken);\n\n// NVIDIA NIM models (primary for text)\nconst NVIDIA_MODELS = [\n \"meta/llama-3.1-8b-instruct\",\n \"meta/llama-3.1-70b-instruct\",\n \"nvidia/llama-3.1-nemotron-70b-instruct\",\n \"mistralai/mistral-7b-instruct-v0.3\",\n];\n\n// NVIDIA Vision models\nconst NVIDIA_VISION_MODELS = [\n \"meta/llama-3.2-11b-vision-instruct\",\n \"meta/llama-3.2-90b-vision-instruct\",\n \"nvidia/vila\",\n];\n\n// OpenRouter fallback models\nconst OPENROUTER_MODELS = [\n \"meta-llama/llama-3.2-3b-instruct:free\",\n \"nousresearch/hermes-3-llama-3.1-405b:free\",\n \"arcee-ai/trinity-mini:free\",\n];\n\n// OpenRouter vision fallback models\nconst OPENROUTER_VISION_MODELS = [\n \"opengvlab/internvl3-2b:free\",\n \"meta-llama/llama-3.2-11b-vision-instruct:free\",\n];\n\n// Poe vision models (new API - OpenAI compatible)\nconst POE_VISION_MODELS = [\n \"Gemini-2.5-Flash-Lite\",\n \"GPT-4o-Mini\",\n \"Grok-4-fast-non-reasoning\"\n];\n\nconst POE_IMAGE_BOT = \"Flux-1-Schnell-FW\";\n\nconst NVIDIA_DOCUMENT_MODELS = [\n \"nvidia/llama-3.1-nemotron-70b-instruct\",\n \"meta/llama-3.1-70b-instruct\",\n];\n\nconst OPENROUTER_DOCUMENT_MODELS = [\n \"tngtech/deepseek-r1t2-chimera:free\",\n \"tngtech/deepseek-r1t-chimera:free\",\n];\n\n// ============ POE API (NEW OpenAI-compatible format) ============\n\nasync function callPoeApi(\n model: string,\n messages: Array<{ role: string; content: unknown }>,\n maxTokens: number = 1024\n): Promise {\n if (!poeApiKey) return null;\n\n try {\n console.log(`Poe API attempt: ${model}`);\n const response = await fetch(\"https://api.poe.com/v1/chat/completions\", {\n method: \"POST\",\n headers: {\n Authorization: `Bearer ${poeApiKey}`,\n \"Content-Type\": \"application/json\",\n },\n body: JSON.stringify({\n model,\n messages,\n max_tokens: maxTokens,\n stream: false,\n }),\n });\n\n if (!response.ok) {\n const errorText = await response.text();\n console.error(`Poe ${model} failed: ${response.status} - ${errorText}`);\n return null;\n }\n\n const data = await response.json();\n const content = data.choices?.[0]?.message?.content;\n if (content?.trim()) {\n console.log(`Poe success with ${model}`);\n return content;\n }\n return null;\n } catch (error) {\n console.error(`Poe ${model} error:`, error);\n return null;\n }\n}\n\n// ============ NVIDIA API CALLS ============\n\nasync function callNvidiaApi(\n model: string,\n messages: Array<{ role: string; content: unknown }>,\n maxTokens: number = 1024\n): Promise {\n if (!nvidiaApiKey) return null;\n\n try {\n console.log(`NVIDIA API attempt: ${model}`);\n const response = await fetch(\n \"https://integrate.api.nvidia.com/v1/chat/completions\",\n {\n method: \"POST\",\n headers: {\n Authorization: `Bearer ${nvidiaApiKey}`,\n \"Content-Type\": \"application/json\",\n },\n body: JSON.stringify({\n model,\n messages,\n max_tokens: maxTokens,\n temperature: 0.7,\n top_p: 0.9,\n }),\n }\n );\n\n if (!response.ok) {\n console.error(`NVIDIA ${model} failed: ${response.status}`);\n return null;\n }\n\n const data = await response.json();\n const content = data.choices?.[0]?.message?.content;\n if (content?.trim()) {\n console.log(`NVIDIA success with ${model}`);\n return content;\n }\n return null;\n } catch (error) {\n console.error(`NVIDIA ${model} error:`, error);\n return null;\n }\n}\n\nasync function callOpenRouterApi(\n model: string,\n messages: Array<{ role: string; content: unknown }>,\n apiKey: string,\n maxTokens: number = 1024\n): Promise {\n try {\n console.log(`OpenRouter API attempt: ${model}`);\n const response = await fetch(\n \"https://openrouter.ai/api/v1/chat/completions\",\n {\n method: \"POST\",\n headers: {\n Authorization: `Bearer ${apiKey}`,\n \"Content-Type\": \"application/json\",\n },\n body: JSON.stringify({\n model,\n messages,\n max_tokens: maxTokens,\n }),\n }\n );\n\n if (!response.ok) {\n console.error(`OpenRouter ${model} failed: ${response.status}`);\n return null;\n }\n\n const data = await response.json();\n const content = data.choices?.[0]?.message?.content;\n if (content?.trim()) {\n console.log(`OpenRouter success with ${model}`);\n return content;\n }\n return null;\n } catch (error) {\n console.error(`OpenRouter ${model} error:`, error);\n return null;\n }\n}\n\n// ============ TEXT GENERATION ============\n\nasync function generateText(\n message: string,\n systemPrompt: string = \"You are a chat user. Answer concise.\"\n): Promise {\n const messages = [\n { role: \"system\", content: systemPrompt },\n { role: \"user\", content: message },\n ];\n\n // Try NVIDIA first\n for (const model of NVIDIA_MODELS) {\n const result = await callNvidiaApi(model, messages);\n if (result) return result;\n }\n\n // Fallback to OpenRouter\n const apiKeys = [openrouterApiKey, openrouterApiKey2].filter(\n Boolean\n ) as string[];\n\n for (const apiKey of apiKeys) {\n for (const model of OPENROUTER_MODELS) {\n const result = await callOpenRouterApi(model, messages, apiKey);\n if (result) return result;\n }\n }\n\n throw new Error(\"All text generation models failed\");\n}\n\n// ============ VISION (IMAGE ANALYSIS) ============\n\nasync function analyzeImageWithPoe(\n imageBase64: string,\n prompt: string\n): Promise {\n if (!poeApiKey) return null;\n\n for (const model of POE_VISION_MODELS) {\n try {\n console.log(`Poe vision attempt: ${model}`);\n\n const messages = [\n {\n role: \"user\",\n content: [\n { type: \"text\", text: prompt },\n {\n type: \"image_url\",\n image_url: { url: `data:image/jpeg;base64,${imageBase64}` },\n },\n ],\n },\n ];\n\n const result = await callPoeApi(model, messages, 1024);\n if (result) return result;\n } catch (error) {\n console.error(`Poe vision ${model} error:`, error);\n }\n }\n\n return null;\n}\n\nasync function tryNvidiaVision(\n model: string,\n imageBase64: string,\n prompt: string\n): Promise {\n if (!nvidiaApiKey) return null;\n\n try {\n console.log(`NVIDIA Vision attempt: ${model}`);\n const messages = [\n {\n role: \"user\",\n content: [\n { type: \"text\", text: prompt },\n {\n type: \"image_url\",\n image_url: { url: `data:image/jpeg;base64,${imageBase64}` },\n },\n ],\n },\n ];\n\n return await callNvidiaApi(model, messages, 1024);\n } catch (error) {\n console.error(`NVIDIA Vision ${model} error:`, error);\n return null;\n }\n}\n\nasync function tryOpenRouterVision(\n model: string,\n imageSource: string,\n prompt: string,\n apiKey: string\n): Promise {\n try {\n console.log(`OpenRouter Vision attempt: ${model}`);\n const messages = [\n {\n role: \"user\",\n content: [\n { type: \"text\", text: prompt },\n { type: \"image_url\", image_url: { url: imageSource } },\n ],\n },\n ];\n\n return await callOpenRouterApi(model, messages, apiKey, 1024);\n } catch (error) {\n console.error(`OpenRouter Vision ${model} error:`, error);\n return null;\n }\n}\n\n/**\n * Vision pipeline:\n * 1) Poe Vision models (primary - new OpenAI-compatible API)\n * 2) NVIDIA Vision models (fallback)\n * 3) OpenRouter Vision models (final fallback)\n */\nasync function generateWithVision(\n imageBase64: string,\n imageUrl: string,\n prompt: string\n): Promise {\n // 1. Try Poe Vision first (new OpenAI-compatible API)\n console.log(\"Trying Poe vision models first...\");\n const poeResult = await analyzeImageWithPoe(imageBase64, prompt);\n if (poeResult) return poeResult;\n\n // 2. Try NVIDIA Vision models\n console.log(\"Poe failed, trying NVIDIA vision...\");\n for (const model of NVIDIA_VISION_MODELS) {\n const result = await tryNvidiaVision(model, imageBase64, prompt);\n if (result) return result;\n }\n\n // 3. Try OpenRouter Vision models\n console.log(\"NVIDIA failed, trying OpenRouter vision...\");\n const apiKeys = [openrouterApiKey, openrouterApiKey2].filter(\n Boolean\n ) as string[];\n\n for (const apiKey of apiKeys) {\n for (const model of OPENROUTER_VISION_MODELS) {\n // Try URL first\n const urlResult = await tryOpenRouterVision(\n model,\n imageUrl,\n prompt,\n apiKey\n );\n if (urlResult) return urlResult;\n\n // Try base64\n const b64Result = await tryOpenRouterVision(\n model,\n `data:image/jpeg;base64,${imageBase64}`,\n prompt,\n apiKey\n );\n if (b64Result) return b64Result;\n }\n }\n\n throw new Error(\"All vision models failed\");\n}\n\n// ============ DOCUMENT PROCESSING ============\n\nasync function extractTextFromDocument(\n fileBase64: string,\n _fileName: string\n): Promise {\n try {\n const decoder = new TextDecoder(\"utf-8\", {\n fatal: false,\n ignoreBOM: true,\n });\n const bytes = Uint8Array.from(atob(fileBase64), (c) => c.charCodeAt(0));\n\n const isPDF =\n bytes[0] === 0x25 &&\n bytes[1] === 0x50 &&\n bytes[2] === 0x44 &&\n bytes[3] === 0x46;\n\n if (isPDF) {\n console.log(\"Detected PDF file - extracting text\");\n const pdfText = decoder.decode(bytes);\n const textChunks: string[] = [];\n\n const btMatches = pdfText.matchAll(/BT\\s+(.*?)\\s+ET/gs);\n for (const match of btMatches) {\n const content = match[1];\n const tjMatches = content.matchAll(/\$(.*?)\$\\s*Tj/g);\n for (const tjMatch of tjMatches) {\n textChunks.push(tjMatch[1]);\n }\n }\n\n if (textChunks.length > 0) {\n return textChunks.join(\" \").trim();\n }\n\n return pdfText\n .replace(/[^\\x20-\\x7E\\n\\r]/g, \" \")\n .replace(/\\s+/g, \" \")\n .trim();\n }\n\n const textContent = decoder.decode(bytes);\n return textContent\n .replace(/\\r\\n/g, \"\\n\")\n .replace(/[^\\x20-\\x7E\\n\\r\\t]/g, \"\")\n .trim();\n } catch (error) {\n console.error(\"Text extraction error:\", error);\n throw new Error(\n `Failed to extract text: ${error instanceof Error ? error.message : \"Unknown error\"}`\n );\n }\n}\n\nasync function processDocument(\n fileBase64: string,\n fileName: string,\n prompt: string\n): Promise {\n const textContent = await extractTextFromDocument(fileBase64, fileName);\n\n if (textContent.length < 10) {\n throw new Error(\"Insufficient text content extracted from document\");\n }\n\n console.log(`Extracted ${textContent.length} characters from ${fileName}`);\n\n const maxLength = 12000;\n const documentContent =\n textContent.length > maxLength\n ? textContent.substring(0, maxLength) + \"\\n\\n[Document truncated...]\"\n : textContent;\n\n const messages = [\n { role: \"system\", content: `Analyze the attached document: \"${fileName}\"` },\n {\n role: \"user\",\n content: `Document content:\\n\\n${documentContent}\\n\\nUser request: ${prompt}`,\n },\n ];\n\n // Try NVIDIA models first\n for (const model of NVIDIA_DOCUMENT_MODELS) {\n const result = await callNvidiaApi(model, messages, 2000);\n if (result) return result;\n }\n\n // Fallback to OpenRouter\n const apiKeys = [openrouterApiKey, openrouterApiKey2].filter(\n Boolean\n ) as string[];\n\n for (const apiKey of apiKeys) {\n for (const model of OPENROUTER_DOCUMENT_MODELS) {\n const result = await callOpenRouterApi(model, messages, apiKey, 2000);\n if (result) return result;\n }\n }\n\n // Return extracted text if all models fail\n return `📄 Document: ${fileName}\\n\\nExtracted text (first 1000 chars):\\n\\n${textContent.substring(0, 1000)}${textContent.length > 1000 ? \"...\" : \"\"}`;\n}\n\n// ============ IMAGE GENERATION ============\n\nasync function generateImageWithPoe(prompt: string): Promise {\n if (!poeApiKey) return null;\n\n try {\n console.log(`Poe image gen: ${POE_IMAGE_BOT}`);\n\n // For image generation, we use the chat completions API\n // The bot will return an image attachment\n const response = await fetch(\"https://api.poe.com/v1/chat/completions\", {\n method: \"POST\",\n headers: {\n Authorization: `Bearer ${poeApiKey}`,\n \"Content-Type\": \"application/json\",\n },\n body: JSON.stringify({\n model: POE_IMAGE_BOT,\n messages: [{ role: \"user\", content: prompt }],\n stream: false,\n }),\n });\n\n if (!response.ok) {\n const errorText = await response.text();\n console.error(`Poe ${POE_IMAGE_BOT} failed: ${response.status} - ${errorText}`);\n return null;\n }\n\n const data = await response.json();\n\n // Check for image in the response\n // The new API may return images in different ways\n const content = data.choices?.[0]?.message?.content;\n\n // Try to extract image URL from markdown\n if (content) {\n const imgMatch = content.match(/!\\[.*?\\]\$(https?:\\/\\/[^\\s)]+)\$/);\n if (imgMatch) {\n console.log(`Poe ${POE_IMAGE_BOT} generated image successfully`);\n return imgMatch[1];\n }\n\n // Try to find direct URL\n const urlMatch = content.match(/(https?:\\/\\/[^\\s]+\\.(png|jpg|jpeg|webp|gif))/i);\n if (urlMatch) {\n console.log(`Poe ${POE_IMAGE_BOT} generated image successfully`);\n return urlMatch[1];\n }\n }\n\n console.log(`Poe ${POE_IMAGE_BOT} returned no image`);\n return null;\n } catch (error) {\n console.error(`Poe ${POE_IMAGE_BOT} error:`, error);\n return null;\n }\n}\n\nasync function generateImageWithPollinations(\n prompt: string\n): Promise {\n try {\n console.log(\"Trying Pollinations.ai\");\n const encodedPrompt = encodeURIComponent(prompt);\n const imageUrl = `https://image.pollinations.ai/prompt/${encodedPrompt}?width=1024&height=1024&nologo=true`;\n\n const checkResponse = await fetch(imageUrl, { method: \"HEAD\" });\n if (checkResponse.ok) {\n console.log(\"Pollinations.ai success\");\n return imageUrl;\n }\n return null;\n } catch (error) {\n console.error(\"Pollinations.ai error:\", error);\n return null;\n }\n}\n\nasync function generateImage(\n prompt: string\n): Promise<{ url: string; isBase64: boolean } | null> {\n const poeUrl = await generateImageWithPoe(prompt);\n if (poeUrl) return { url: poeUrl, isBase64: false };\n\n const pollinationsUrl = await generateImageWithPollinations(prompt);\n if (pollinationsUrl) return { url: pollinationsUrl, isBase64: false };\n\n return null;\n}\n\n// ============ FILE DOWNLOAD ============\n\nasync function downloadFile(\n fileId: string\n): Promise<{ data: Uint8Array; url: string }> {\n try {\n const file = await bot.api.getFile(fileId);\n const tgUrl = `https://api.telegram.org/file/bot${telegramToken}/${file.file_path}`;\n const response = await fetch(tgUrl);\n if (!response.ok) {\n throw new Error(`File download failed: ${response.status}`);\n }\n const data = new Uint8Array(await response.arrayBuffer());\n\n let publicUrl = tgUrl;\n try {\n const formData = new FormData();\n formData.append(\n \"file\",\n new Blob([data], { type: \"image/jpeg\" }),\n \"image.jpg\"\n );\n const uploadResp = await fetch(\"https://tmpfiles.org/api/v1/upload\", {\n method: \"POST\",\n body: formData,\n });\n if (uploadResp.ok) {\n const uploadData = await uploadResp.json();\n if (uploadData?.data?.url) {\n publicUrl = uploadData.data.url.replace(\n \"tmpfiles.org/\",\n \"tmpfiles.org/dl/\"\n );\n console.log(`Uploaded to public URL: ${publicUrl}`);\n }\n }\n } catch (uploadErr) {\n console.log(\"Public upload failed, using TG URL:\", uploadErr);\n }\n\n return { data, url: publicUrl };\n } catch (error) {\n console.error(\"Download error:\", error);\n throw error;\n }\n}\n\n// ============ AUTO SEARCH ============\n\ninterface AutoSearch {\n chatId: number;\n query: string;\n lastRun: number;\n}\n\nconst autoSearches = new Map();\n\nasync function performWebSearch(query: string): Promise {\n try {\n return await generateText(query, \"You are a chat user. Answer concise.\");\n } catch (error) {\n console.error(\"Search error:\", error);\n return \"Search unavailable.\";\n }\n}\n\nasync function runAutoSearches() {\n const now = Date.now();\n const eightHours = 8 * 60 * 60 * 1000;\n\n for (const [key, search] of autoSearches.entries()) {\n if (now - search.lastRun >= eightHours) {\n try {\n const result = await performWebSearch(search.query);\n await bot.api.sendMessage(\n search.chatId,\n `🔍 Auto Search: \"${search.query}\"\\n\\n${result}`\n );\n search.lastRun = now;\n autoSearches.set(key, search);\n } catch (error) {\n console.error(`Auto search failed for ${search.chatId}:`, error);\n }\n }\n }\n}\n\nsetInterval(runAutoSearches, 30 * 60 * 1000);\n\n// ============ HELPERS ============\n\nfunction isImageGenerationRequest(text: string): boolean {\n if (!text) return false;\n const lower = text.toLowerCase().trim();\n const keywords = [\n \"generate image\",\n \"gpt photo\",\n \"create image\",\n \"image of\",\n \"photo of\",\n \"draw\",\n \"illustrate\",\n \"/image\",\n \"--image\",\n ];\n return keywords.some((k) => lower.startsWith(k));\n}\n\n// ============ BOT COMMANDS ============\n\nbot.command(\"start\", (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup) {\n return ctx.reply(\"Join group: https://t.me/chatgpt_deutsch\");\n }\n return ctx.reply(\"Poe AI access: https://poe.com/Open-Hat-Lab\");\n});\n\nbot.command(\"search\", async (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup)\n return ctx.reply(\"Group only: https://t.me/chatgpt_deutsch\");\n\n const query = ctx.message?.text?.replace(\"/search\", \"\").trim();\n if (!query) return ctx.reply(\"Usage: /search \");\n\n try {\n await ctx.replyWithChatAction(\"typing\");\n const response = await generateText(\n query,\n \"You are a chat user. Answer concise.\"\n );\n await ctx.reply(response);\n } catch (error) {\n console.error(\"Search error:\", error);\n await ctx.reply(\"Search failed. Try: https://poe.com/Open-Hat-Lab\");\n }\n});\n\nbot.command(\"auto_search\", async (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup) return ctx.reply(\"Group only\");\n\n const query = ctx.message?.text?.replace(\"/auto_search\", \"\").trim();\n if (!query) return ctx.reply(\"Usage: /auto_search \");\n\n const chatId = ctx.chat.id;\n const key = `${chatId}:${query}`;\n autoSearches.set(key, { chatId, query, lastRun: Date.now() });\n\n await ctx.replyWithChatAction(\"typing\");\n const result = await performWebSearch(query);\n await ctx.reply(`✅ Auto search: \"${query}\"\\nUpdates every 8h.\\n\\n${result}`);\n});\n\nbot.command(\"list_searches\", (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup) return ctx.reply(\"Group only\");\n\n const searches = Array.from(autoSearches.values()).filter(\n (s) => s.chatId === ctx.chat.id\n );\n if (searches.length === 0) return ctx.reply(\"No active searches\");\n\n const list = searches.map((s, i) => `${i + 1}. \"${s.query}\"`).join(\"\\n\");\n return ctx.reply(`Active searches:\\n\\n${list}`);\n});\n\nbot.command(\"stop_search\", (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup) return ctx.reply(\"Group only\");\n\n const query = ctx.message?.text?.replace(\"/stop_search\", \"\").trim();\n if (!query) return ctx.reply(\"Usage: /stop_search \");\n\n const key = `${ctx.chat.id}:${query}`;\n if (autoSearches.delete(key)) {\n return ctx.reply(`✅ Stopped: \"${query}\"`);\n }\n return ctx.reply(`❌ Not found: \"${query}\"`);\n});\n\nbot.command(\"image\", async (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup)\n return ctx.reply(\"Group only: https://t.me/chatgpt_deutsch\");\n\n const prompt = ctx.message?.text?.replace(\"/image\", \"\").trim();\n if (!prompt || prompt.length < 3) {\n return ctx.reply(\"Usage: /image \");\n }\n\n try {\n await ctx.replyWithChatAction(\"upload_photo\");\n const result = await generateImage(prompt);\n\n if (result) {\n if (result.isBase64) {\n const base64Data = result.url.split(\",\")[1];\n const imageBuffer = Uint8Array.from(atob(base64Data), (c) =>\n c.charCodeAt(0)\n );\n await ctx.replyWithPhoto(new InputFile(imageBuffer, \"image.png\"), {\n caption: `🎨 \"${prompt}\"`,\n });\n } else {\n await ctx.replyWithPhoto(result.url, { caption: `🎨 \"${prompt}\"` });\n }\n } else {\n await ctx.reply(\"Image generation failed.\");\n }\n } catch (error) {\n console.error(\"Image error:\", error);\n await ctx.reply(\n `Generation error: ${error instanceof Error ? error.message : \"Unknown\"}`\n );\n }\n});\n\n// ============ MESSAGE HANDLER ============\n\nbot.on(\"message\", async (ctx) => {\n const isGroup =\n ctx.chat?.type === \"group\" || ctx.chat?.type === \"supergroup\";\n if (!isGroup) {\n return ctx.reply(\"Join: https://t.me/chatgpt_deutsch\");\n }\n\n const messageText = ctx.message?.text || ctx.message?.caption || \"\";\n const photo = ctx.message?.photo;\n const document = ctx.message?.document;\n\n // --- Image generation from text ---\n if (isImageGenerationRequest(messageText)) {\n let prompt = messageText;\n for (const prefix of [\n \"generate image\",\n \"gpt photo\",\n \"create image\",\n \"image of\",\n \"/image\",\n ]) {\n if (prompt.toLowerCase().startsWith(prefix)) {\n prompt = prompt.substring(prefix.length).trim();\n break;\n }\n }\n if (prompt.length < 3) return ctx.reply(\"Provide image description\");\n\n try {\n await ctx.replyWithChatAction(\"upload_photo\");\n const result = await generateImage(prompt);\n\n if (result) {\n if (result.isBase64) {\n const base64Data = result.url.split(\",\")[1];\n const imageBuffer = Uint8Array.from(atob(base64Data), (c) =>\n c.charCodeAt(0)\n );\n await ctx.replyWithPhoto(new InputFile(imageBuffer, \"image.png\"), {\n caption: `🎨 \"${prompt}\"`,\n });\n } else {\n await ctx.replyWithPhoto(result.url, { caption: `🎨 \"${prompt}\"` });\n }\n } else {\n await ctx.reply(\"All image providers unavailable.\");\n }\n } catch (error) {\n console.error(\"Auto image error:\", error);\n await ctx.reply(\"Use /image \");\n }\n return;\n }\n\n // --- Image analysis (photo sent) ---\n if (photo) {\n try {\n await ctx.replyWithChatAction(\"typing\");\n const largestPhoto = photo[photo.length - 1];\n console.log(`Processing photo: ${largestPhoto.file_id}`);\n\n const { data: imageData, url: imageUrl } = await downloadFile(\n largestPhoto.file_id\n );\n\n let base64 = \"\";\n const chunkSize = 8192;\n for (let i = 0; i < imageData.length; i += chunkSize) {\n const chunk = imageData.subarray(i, i + chunkSize);\n base64 += btoa(String.fromCharCode(...chunk));\n }\n\n const prompt = messageText || \"Describe this image concisely\";\n\n console.log(`Image size: ${imageData.length} bytes`);\n\n const response = await generateWithVision(base64, imageUrl, prompt);\n await ctx.reply(response);\n } catch (error) {\n console.error(\"Vision error:\", error);\n await ctx.reply(\n `Image analysis failed: ${error instanceof Error ? error.message : \"Unknown error\"}`\n );\n }\n return;\n }\n\n // --- Document processing ---\n if (document) {\n try {\n await ctx.replyWithChatAction(\"typing\");\n const fileName = document.file_name || \"document\";\n console.log(\n `Processing document: ${fileName}, size: ${document.file_size} bytes`\n );\n\n if (document.file_size && document.file_size > 20 * 1024 * 1024) {\n return ctx.reply(\"Maximum size: 20MB\");\n }\n\n const { data: fileData } = await downloadFile(document.file_id);\n const base64 = btoa(String.fromCharCode(...fileData));\n const prompt = messageText || \"Summarize this document.\";\n\n console.log(\n `Document downloaded: ${fileData.length} bytes, prompt: \"${prompt}\"`\n );\n\n const response = await processDocument(base64, fileName, prompt);\n await ctx.reply(response);\n } catch (error) {\n console.error(\"Document error:\", error);\n await ctx.reply(\n `Document processing failed: ${error instanceof Error ? error.message : \"Unknown error\"}`\n );\n }\n return;\n }\n});\n\n// ============ ERROR HANDLER & WEBHOOK ============\n\nbot.catch((err) => {\n console.error(\"Bot error:\", err);\n});\n\nconst SECRET_TOKEN =\n Deno.env.get(\"OPENROUTER_API_KEY\") || crypto.randomUUID();\nconst handleUpdate = webhookCallback(\n bot,\n \"std/http\",\n undefined,\n undefined,\n SECRET_TOKEN\n);\n\nexport default async (req: Request): Promise => {\n try {\n const url = new URL(req.url);\n const webhookUrl = `${url.protocol}//${url.host}${url.pathname}`;\n await bot.api.setWebhook(webhookUrl, { secret_token: SECRET_TOKEN });\n } catch (error) {\n console.error(\"Webhook setup error:\", error);\n }\n\n if (req.method === \"POST\") {\n try {\n return await handleUpdate(req);\n } catch (error) {\n console.error(\"Update error:\", error);\n return new Response(\"OK\", { status: 200 });\n }\n }\n\n return new Response(\"Bot active\", {\n status: 200,\n headers: { \"content-type\": \"text/html\" },\n });\n};","2026-02-10T14:38:02.790Z","Free-Omni-Telegram","https://www.val.town/x/RobooHood/Free-Omni-Telegram/code/","RobooHood","https://poe.com/Cheap-Omni","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/84461d95-36d3-4783-01d6-1d1eeda8a100/square512",[204,207,209,212,215,217,220,222,225,228,231,233,235,238,240,243,246,249,252,255,258,261,264,267,270],{"_129":205,"_131":206,"_133":22},17,"const openrouterApiKey2 = Deno.env.get(\"OPENROUTER_API_KEY2\");",{"_129":208,"_131":74,"_133":22},18,{"_129":210,"_131":211,"_133":139},19,"// Poe API (new OpenAI-compatible format)",{"_129":213,"_131":214,"_133":22},20,"const poeApiKey = Deno.env.get(\"POE_API_KEY\");",{"_129":216,"_131":74,"_133":22},21,{"_129":218,"_131":219,"_133":22},50,"];",{"_129":221,"_131":74,"_133":22},51,{"_129":223,"_131":224,"_133":139},52,"// Poe vision models (new API - OpenAI compatible)",{"_129":226,"_131":227,"_133":22},53,"const POE_VISION_MODELS = [",{"_129":229,"_131":230,"_133":22},54," \"Gemini-2.5-Flash-Lite\",",{"_129":232,"_131":219,"_133":22},69,{"_129":234,"_131":74,"_133":22},70,{"_129":236,"_131":237,"_133":139},71,"// ============ POE API (NEW OpenAI-compatible format) ============",{"_129":239,"_131":74,"_133":22},72,{"_129":241,"_131":242,"_133":22},73,"async function callPoeApi(",{"_129":244,"_131":245,"_133":22},325,"/**",{"_129":247,"_131":248,"_133":22},326," * Vision pipeline:",{"_129":250,"_131":251,"_133":139},327," * 1) Poe Vision models (primary - new OpenAI-compatible API)",{"_129":253,"_131":254,"_133":22},328," * 2) NVIDIA Vision models (fallback)",{"_129":256,"_131":257,"_133":22},329," * 3) OpenRouter Vision models (final fallback)",{"_129":259,"_131":260,"_133":22},334," prompt: string",{"_129":262,"_131":263,"_133":22},335,"): Promise {",{"_129":265,"_131":266,"_133":139},336," // 1. Try Poe Vision first (new OpenAI-compatible API)",{"_129":268,"_131":269,"_133":22},337," console.log(\"Trying Poe vision models first...\");",{"_129":271,"_131":272,"_133":22},338," const poeResult = await analyzeImageWithPoe(imageBase64, prompt);",5,{"_106":275,"_108":276,"_110":277,"_112":278,"_114":279,"_116":280,"_118":281,"_120":282,"_122":283,"_124":-5,"_125":-5,"_126":284,"_160":368},"fc87369c-056f-11f1-9b9e-42dde27851f2","https://www.val.town/x/mayank_yadav/FinID/code/main.ts","fc85282a-056f-11f1-9b9e-42dde27851f2","main.ts","import { OpenAI } from \"https://esm.sh/openai@4.20.1\";\n\nexport default async function handler(req: Request): Promise {\n const corsHeaders = {\n \"Access-Control-Allow-Origin\": \"*\",\n \"Access-Control-Allow-Methods\": \"POST, OPTIONS\",\n \"Access-Control-Allow-Headers\": \"Content-Type, Authorization\",\n };\n\n if (req.method === \"OPTIONS\") {\n return new Response(null, { status: 204, headers: corsHeaders });\n }\n\n // Mandatory authentication check\n const authHeader = req.headers.get(\"Authorization\");\n const expectedSecret = Deno.env.get(\"IOS_APP_SECRET\");\n\n if (!authHeader || !authHeader.startsWith(\"Bearer \")) {\n return new Response(\n JSON.stringify({ success: false, error: \"Authentication required\" }),\n {\n status: 401,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n const providedSecret = authHeader.replace(\"Bearer \", \"\");\n if (expectedSecret && providedSecret !== expectedSecret) {\n return new Response(\n JSON.stringify({ success: false, error: \"Invalid authentication\" }),\n {\n status: 401,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n console.log(\"✅ Authenticated request\");\n\n if (req.method !== \"POST\") {\n return new Response(\n JSON.stringify({ success: false, error: \"Method not allowed\" }),\n {\n status: 405,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n try {\n const body = await req.json();\n const { imageBase64, images, language } = body;\n\n // Support both old format (single image) and new format (multiple images)\n const base64Images = images || (imageBase64 ? [imageBase64] : []);\n\n if (base64Images.length === 0) {\n return new Response(\n JSON.stringify({\n success: false,\n error: \"Missing image data in request body\",\n }),\n {\n status: 400,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n // Get language preference (default to English)\n const userLanguage = language || \"English\";\n \n console.log(`🌍 User language: ${userLanguage}`);\n console.log(`📸 Processing ${base64Images.length} image(s)`);\n\n const apiKey = Deno.env.get(\"OPENAI_API_KEY\");\n if (!apiKey) {\n console.error(\"❌ OPENAI_API_KEY not set\");\n return new Response(\n JSON.stringify({\n success: false,\n error: \"Something went wrong. Please try again later.\",\n }),\n {\n status: 500,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n const openai = new OpenAI({ apiKey });\n\n console.log(\"🤖 Calling OpenAI for fish identification...\");\n\n // Language-specific instruction\n const languageInstruction = userLanguage !== \"English\"\n ? `\\n\\nIMPORTANT: For all text fields (except enum fields), respond in ${userLanguage}. All descriptive text must be in ${userLanguage}.`\n : \"\";\n\n // Enum instruction - CRITICAL for app logic\n const enumInstruction = `\\n\\nCRITICAL - DUAL FIELD SYSTEM:\nSome fields have TWO versions - an enum version (always English) and a display version (translated to ${userLanguage}):\n\nENUM FIELDS (use EXACT English values, do NOT translate these):\n- edibilityStatusEnum: MUST be one of: \"edible\", \"not_edible\", \"not_recommended\", \"toxic\"\n- temperamentEnum: MUST be one of: \"peaceful\", \"semi_aggressive\", \"aggressive\"\n- socialBehaviorEnum: MUST be one of: \"schooling\", \"solitary\", \"pairs\"\n- difficultyLevelEnum: MUST be one of: \"beginner\", \"intermediate\", \"advanced\", \"expert\"\n- dietTypeEnum: MUST be one of: \"carnivore\", \"herbivore\", \"omnivore\"\n- breedingDifficultyEnum: MUST be one of: \"easy\", \"moderate\", \"difficult\", \"very_difficult\"\n- tankTypeEnum: MUST be one of: \"freshwater\", \"saltwater\", \"brackish\"\n\nDISPLAY FIELDS (translate to ${userLanguage} for user display):\n- edibilityStatus: User-friendly translated version of edibilityStatusEnum\n- temperament: User-friendly translated version of temperamentEnum\n- socialBehavior: User-friendly translated version of socialBehaviorEnum\n- difficultyLevel: User-friendly translated version of difficultyLevelEnum\n- dietType: User-friendly translated version of dietTypeEnum\n- breedingDifficulty: User-friendly translated version of breedingDifficultyEnum\n- tankType: User-friendly translated version of tankTypeEnum\n\nExample for Japanese user:\n{\n \"edibilityStatusEnum\": \"edible\", // ← ALWAYS English enum\n \"edibilityStatus\": \"食用\", // ← Translated for display\n \"temperamentEnum\": \"peaceful\", // ← ALWAYS English enum\n \"temperament\": \"温和\" // ← Translated for display\n}`;\n\n // Multilingual error messages for multiple fish detection\n const multipleFishErrors: Record = {\n \"English\": \"Please identify one fish at a time for more accurate results. The image appears to contain multiple fish.\",\n \"Japanese\": \"より正確な結果を得るため、一度に1匹の魚を識別してください。画像に複数の魚が含まれているようです。\",\n \"Spanish\": \"Por favor, identifique un pez a la vez para obtener resultados más precisos. La imagen parece contener múltiples peces.\",\n \"German\": \"Bitte identifizieren Sie jeweils einen Fisch für genauere Ergebnisse. Das Bild scheint mehrere Fische zu enthalten.\",\n };\n\n const multipleFishError = multipleFishErrors[userLanguage] || multipleFishErrors[\"English\"];\n\n // Build content array with text prompt + all images\n const contentArray = [\n {\n type: \"text\",\n text: base64Images.length > 1\n ? `You are an expert ichthyologist and marine biologist with deep knowledge of fish species, taxonomy, habitats, and identification. Analyze these ${base64Images.length} images of the SAME fish from different angles.\n\nCRITICAL VALIDATION RULES:\n1. All images MUST show the SAME fish from different angles/perspectives\n2. If images show DIFFERENT fish, respond with the multiple fish error\n3. If you detect MORE THAN ONE fish in any single image, respond with the multiple fish error\n\nIf validation fails, respond with:\n{\n \"error\": \"${multipleFishError}\",\n \"commonName\": null,\n \"scientificName\": null,\n \"alternativeNames\": null,\n \"family\": null,\n \"order\": null,\n \"averageLength\": null,\n \"averageWeight\": null,\n \"maximumLength\": null,\n \"maximumWeight\": null,\n \"distinctiveFeatures\": null,\n \"colorVariations\": null,\n \"sexualDimorphism\": null,\n \"nativeRange\": null,\n \"habitatType\": null,\n \"preferredDepth\": null,\n \"temperatureRange\": null,\n \"commonLocations\": null,\n \"edibilityStatusEnum\": null,\n \"edibilityStatus\": null,\n \"tasteDescription\": null,\n \"cookingMethods\": null,\n \"mercuryConcerns\": null,\n \"toxicityWarnings\": null,\n \"dangerousFeatures\": null,\n \"handlingPrecautions\": null,\n \"fishingMethods\": null,\n \"bestBaitLures\": null,\n \"bestSeasons\": null,\n \"bestTimes\": null,\n \"fightingCharacteristics\": null,\n \"isPopularSportFish\": false,\n \"conservationStatus\": null,\n \"isProtected\": false,\n \"isInvasive\": false,\n \"similarSpecies\": null,\n \"identificationDifferences\": null,\n \"isCommonPetFish\": false,\n \"minimumTankSize\": null,\n \"tankTypeEnum\": null,\n \"tankType\": null,\n \"waterTemperatureRange\": null,\n \"phRange\": null,\n \"waterHardness\": null,\n \"filtrationRequirements\": null,\n \"specialEquipment\": null,\n \"temperamentEnum\": null,\n \"temperament\": null,\n \"socialBehaviorEnum\": null,\n \"socialBehavior\": null,\n \"compatibleTankMates\": null,\n \"incompatibleTankMates\": null,\n \"isReefSafe\": false,\n \"difficultyLevelEnum\": null,\n \"difficultyLevel\": null,\n \"dietTypeEnum\": null,\n \"dietType\": null,\n \"foodRecommendations\": null,\n \"feedingFrequency\": null,\n \"lifespanInCaptivity\": null,\n \"breedingDifficultyEnum\": null,\n \"breedingDifficulty\": null,\n \"breedingRequirements\": null,\n \"sexualDifferences\": null,\n \"commonDiseases\": null,\n \"diseaseSusceptibility\": null,\n \"specialCareConsiderations\": null,\n \"confidence\": 0.0\n}\n\nIf all images show the SAME SINGLE fish, use details from all angles to provide comprehensive identification.\n\nRespond with ONLY valid JSON (no markdown, no code blocks, no extra text):\n\n{\n \"commonName\": \"Common name in ${userLanguage}\",\n \"scientificName\": \"Scientific name (Genus species) - NEVER translate\",\n \"alternativeNames\": \"Other common names in ${userLanguage}, separated by commas\" or null,\n \"family\": \"Taxonomic family in ${userLanguage}\",\n \"order\": \"Taxonomic order in ${userLanguage}\",\n \n \"averageLength\": \"Average adult length in ${userLanguage} (e.g., '30-40 cm', '12-16 inches')\",\n \"averageWeight\": \"Average adult weight in ${userLanguage} (e.g., '2-3 kg', '4-7 lbs')\" or null,\n \"maximumLength\": \"Maximum recorded length in ${userLanguage}\" or null,\n \"maximumWeight\": \"Maximum recorded weight in ${userLanguage}\" or null,\n \"distinctiveFeatures\": \"2-3 sentences in ${userLanguage} describing key identifying features\",\n \"colorVariations\": \"Description in ${userLanguage} of color variations\" or null,\n \"sexualDimorphism\": \"Differences between males and females in ${userLanguage}\" or null,\n \n \"nativeRange\": \"Geographic distribution in ${userLanguage}\",\n \"habitatType\": \"Primary habitat in ${userLanguage}\",\n \"preferredDepth\": \"Depth range in ${userLanguage}\" or null,\n \"temperatureRange\": \"Preferred water temperature in ${userLanguage}\" or null,\n \"commonLocations\": \"Specific regions in ${userLanguage}\" or null,\n \n \"edibilityStatusEnum\": \"edible\" or \"not_edible\" or \"not_recommended\" or \"toxic\" or null (EXACT English enum),\n \"edibilityStatus\": \"Translated edibility status in ${userLanguage}\" or null,\n \"tasteDescription\": \"Taste and texture in ${userLanguage} if edible\" or null,\n \"cookingMethods\": \"Common cooking methods in ${userLanguage} if edible\" or null,\n \"mercuryConcerns\": \"Mercury level concerns in ${userLanguage} if edible\" or null,\n \"toxicityWarnings\": \"Specific toxicity warnings in ${userLanguage}\" or null,\n \"dangerousFeatures\": \"Description of dangerous features in ${userLanguage}\" or null,\n \"handlingPrecautions\": \"Safety precautions in ${userLanguage}\" or null,\n \n \"fishingMethods\": \"Common fishing techniques in ${userLanguage}\" or null,\n \"bestBaitLures\": \"Effective baits or lures in ${userLanguage}\" or null,\n \"bestSeasons\": \"Best fishing seasons in ${userLanguage}\" or null,\n \"bestTimes\": \"Best times of day in ${userLanguage}\" or null,\n \"fightingCharacteristics\": \"Fighting behavior in ${userLanguage}\" or null,\n \"isPopularSportFish\": true/false,\n \n \"conservationStatus\": \"IUCN status or conservation notes in ${userLanguage}\" or null,\n \"isProtected\": true/false,\n \"isInvasive\": true/false,\n \"similarSpecies\": \"List of similar-looking species in ${userLanguage}\" or null,\n \"identificationDifferences\": \"How to distinguish in ${userLanguage}\" or null,\n \n \"isCommonPetFish\": true/false,\n \"minimumTankSize\": \"Minimum aquarium size in ${userLanguage}\" or null,\n \"tankTypeEnum\": \"freshwater\" or \"saltwater\" or \"brackish\" or null (EXACT English enum),\n \"tankType\": \"Translated tank type in ${userLanguage}\" or null,\n \"waterTemperatureRange\": \"Required temperature range in ${userLanguage}\" or null,\n \"phRange\": \"Required pH range in ${userLanguage}\" or null,\n \"waterHardness\": \"Water hardness requirements in ${userLanguage}\" or null,\n \"filtrationRequirements\": \"Filtration needs in ${userLanguage}\" or null,\n \"specialEquipment\": \"Any special equipment in ${userLanguage}\" or null,\n \"temperamentEnum\": \"peaceful\" or \"semi_aggressive\" or \"aggressive\" or null (EXACT English enum),\n \"temperament\": \"Translated temperament in ${userLanguage}\" or null,\n \"socialBehaviorEnum\": \"schooling\" or \"solitary\" or \"pairs\" or null (EXACT English enum),\n \"socialBehavior\": \"Translated social behavior in ${userLanguage}\" or null,\n \"compatibleTankMates\": \"Compatible species in ${userLanguage}\" or null,\n \"incompatibleTankMates\": \"Incompatible species in ${userLanguage}\" or null,\n \"isReefSafe\": true/false,\n \"difficultyLevelEnum\": \"beginner\" or \"intermediate\" or \"advanced\" or \"expert\" or null (EXACT English enum),\n \"difficultyLevel\": \"Translated difficulty level in ${userLanguage}\" or null,\n \"dietTypeEnum\": \"carnivore\" or \"herbivore\" or \"omnivore\" or null (EXACT English enum),\n \"dietType\": \"Translated diet type in ${userLanguage}\" or null,\n \"foodRecommendations\": \"Recommended foods in ${userLanguage}\" or null,\n \"feedingFrequency\": \"How often to feed in ${userLanguage}\" or null,\n \"lifespanInCaptivity\": \"Expected lifespan in ${userLanguage}\" or null,\n \"breedingDifficultyEnum\": \"easy\" or \"moderate\" or \"difficult\" or \"very_difficult\" or null (EXACT English enum),\n \"breedingDifficulty\": \"Translated breeding difficulty in ${userLanguage}\" or null,\n \"breedingRequirements\": \"Conditions needed in ${userLanguage}\" or null,\n \"sexualDifferences\": \"How to distinguish males/females in ${userLanguage}\" or null,\n \"commonDiseases\": \"Common health issues in ${userLanguage}\" or null,\n \"diseaseSusceptibility\": \"Susceptibility to diseases in ${userLanguage}\" or null,\n \"specialCareConsiderations\": \"Any special care notes in ${userLanguage}\" or null,\n \n \"confidence\": 0.0 to 1.0 (your confidence in this identification)\n}\n\nIMPORTANT GUIDELINES:\n- Use information from ALL images to make the most accurate identification\n- Provide detailed, practical information\n- Consider safety heavily - prioritize toxicity/danger warnings\n- If you cannot identify clearly, set confidence below 0.5\n- If you can very confident, set confidence between 0.9 to 0.99 as you see fit\n- Use null for fields you cannot determine with confidence\n- Focus on observable details across all images${languageInstruction}${enumInstruction}`\n : `You are an expert ichthyologist and marine biologist with deep knowledge of fish species, taxonomy, habitats, and identification. Analyze this image and identify the fish.\n\nCRITICAL VALIDATION RULE:\n- If you detect MORE THAN ONE fish in the image, respond with the multiple fish error\n\nIf you detect multiple fish, respond with:\n{\n \"error\": \"${multipleFishError}\",\n \"commonName\": null,\n \"scientificName\": null,\n \"alternativeNames\": null,\n \"family\": null,\n \"order\": null,\n \"averageLength\": null,\n \"averageWeight\": null,\n \"maximumLength\": null,\n \"maximumWeight\": null,\n \"distinctiveFeatures\": null,\n \"colorVariations\": null,\n \"sexualDimorphism\": null,\n \"nativeRange\": null,\n \"habitatType\": null,\n \"preferredDepth\": null,\n \"temperatureRange\": null,\n \"commonLocations\": null,\n \"edibilityStatusEnum\": null,\n \"edibilityStatus\": null,\n \"tasteDescription\": null,\n \"cookingMethods\": null,\n \"mercuryConcerns\": null,\n \"toxicityWarnings\": null,\n \"dangerousFeatures\": null,\n \"handlingPrecautions\": null,\n \"fishingMethods\": null,\n \"bestBaitLures\": null,\n \"bestSeasons\": null,\n \"bestTimes\": null,\n \"fightingCharacteristics\": null,\n \"isPopularSportFish\": false,\n \"conservationStatus\": null,\n \"isProtected\": false,\n \"isInvasive\": false,\n \"similarSpecies\": null,\n \"identificationDifferences\": null,\n \"isCommonPetFish\": false,\n \"minimumTankSize\": null,\n \"tankTypeEnum\": null,\n \"tankType\": null,\n \"waterTemperatureRange\": null,\n \"phRange\": null,\n \"waterHardness\": null,\n \"filtrationRequirements\": null,\n \"specialEquipment\": null,\n \"temperamentEnum\": null,\n \"temperament\": null,\n \"socialBehaviorEnum\": null,\n \"socialBehavior\": null,\n \"compatibleTankMates\": null,\n \"incompatibleTankMates\": null,\n \"isReefSafe\": false,\n \"difficultyLevelEnum\": null,\n \"difficultyLevel\": null,\n \"dietTypeEnum\": null,\n \"dietType\": null,\n \"foodRecommendations\": null,\n \"feedingFrequency\": null,\n \"lifespanInCaptivity\": null,\n \"breedingDifficultyEnum\": null,\n \"breedingDifficulty\": null,\n \"breedingRequirements\": null,\n \"sexualDifferences\": null,\n \"commonDiseases\": null,\n \"diseaseSusceptibility\": null,\n \"specialCareConsiderations\": null,\n \"confidence\": 0.0\n}\n\nIf the image shows a SINGLE fish, respond with ONLY valid JSON (no markdown, no code blocks, no extra text):\n\n{\n \"commonName\": \"Common name in ${userLanguage}\",\n \"scientificName\": \"Scientific name (Genus species) - NEVER translate\",\n \"alternativeNames\": \"Other common names in ${userLanguage}, separated by commas\" or null,\n \"family\": \"Taxonomic family in ${userLanguage}\",\n \"order\": \"Taxonomic order in ${userLanguage}\",\n \n \"averageLength\": \"Average adult length in ${userLanguage}\",\n \"averageWeight\": \"Average adult weight in ${userLanguage}\" or null,\n \"maximumLength\": \"Maximum recorded length in ${userLanguage}\" or null,\n \"maximumWeight\": \"Maximum recorded weight in ${userLanguage}\" or null,\n \"distinctiveFeatures\": \"2-3 sentences in ${userLanguage} describing key identifying features\",\n \"colorVariations\": \"Description in ${userLanguage}\" or null,\n \"sexualDimorphism\": \"Differences in ${userLanguage}\" or null,\n \n \"nativeRange\": \"Geographic distribution in ${userLanguage}\",\n \"habitatType\": \"Primary habitat in ${userLanguage}\",\n \"preferredDepth\": \"Depth range in ${userLanguage}\" or null,\n \"temperatureRange\": \"Preferred water temperature in ${userLanguage}\" or null,\n \"commonLocations\": \"Specific regions in ${userLanguage}\" or null,\n \n \"edibilityStatusEnum\": \"edible\" or \"not_edible\" or \"not_recommended\" or \"toxic\" or null (EXACT English enum),\n \"edibilityStatus\": \"Translated edibility status in ${userLanguage}\" or null,\n \"tasteDescription\": \"Taste description in ${userLanguage}\" or null,\n \"cookingMethods\": \"Cooking methods in ${userLanguage}\" or null,\n \"mercuryConcerns\": \"Mercury concerns in ${userLanguage}\" or null,\n \"toxicityWarnings\": \"Toxicity warnings in ${userLanguage}\" or null,\n \"dangerousFeatures\": \"Dangerous features in ${userLanguage}\" or null,\n \"handlingPrecautions\": \"Safety precautions in ${userLanguage}\" or null,\n \n \"fishingMethods\": \"Fishing techniques in ${userLanguage}\" or null,\n \"bestBaitLures\": \"Baits/lures in ${userLanguage}\" or null,\n \"bestSeasons\": \"Best seasons in ${userLanguage}\" or null,\n \"bestTimes\": \"Best times in ${userLanguage}\" or null,\n \"fightingCharacteristics\": \"Fighting behavior in ${userLanguage}\" or null,\n \"isPopularSportFish\": true/false,\n \n \"conservationStatus\": \"Conservation notes in ${userLanguage}\" or null,\n \"isProtected\": true/false,\n \"isInvasive\": true/false,\n \"similarSpecies\": \"Similar species in ${userLanguage}\" or null,\n \"identificationDifferences\": \"Differences in ${userLanguage}\" or null,\n \n \"isCommonPetFish\": true/false,\n \"minimumTankSize\": \"Tank size in ${userLanguage}\" or null,\n \"tankTypeEnum\": \"freshwater\" or \"saltwater\" or \"brackish\" or null (EXACT English enum),\n \"tankType\": \"Translated tank type in ${userLanguage}\" or null,\n \"waterTemperatureRange\": \"Temperature range in ${userLanguage}\" or null,\n \"phRange\": \"pH range in ${userLanguage}\" or null,\n \"waterHardness\": \"Water hardness in ${userLanguage}\" or null,\n \"filtrationRequirements\": \"Filtration in ${userLanguage}\" or null,\n \"specialEquipment\": \"Special equipment in ${userLanguage}\" or null,\n \"temperamentEnum\": \"peaceful\" or \"semi_aggressive\" or \"aggressive\" or null (EXACT English enum),\n \"temperament\": \"Translated temperament in ${userLanguage}\" or null,\n \"socialBehaviorEnum\": \"schooling\" or \"solitary\" or \"pairs\" or null (EXACT English enum),\n \"socialBehavior\": \"Translated social behavior in ${userLanguage}\" or null,\n \"compatibleTankMates\": \"Compatible species in ${userLanguage}\" or null,\n \"incompatibleTankMates\": \"Incompatible species in ${userLanguage}\" or null,\n \"isReefSafe\": true/false,\n \"difficultyLevelEnum\": \"beginner\" or \"intermediate\" or \"advanced\" or \"expert\" or null (EXACT English enum),\n \"difficultyLevel\": \"Translated difficulty in ${userLanguage}\" or null,\n \"dietTypeEnum\": \"carnivore\" or \"herbivore\" or \"omnivore\" or null (EXACT English enum),\n \"dietType\": \"Translated diet in ${userLanguage}\" or null,\n \"foodRecommendations\": \"Food recommendations in ${userLanguage}\" or null,\n \"feedingFrequency\": \"Feeding frequency in ${userLanguage}\" or null,\n \"lifespanInCaptivity\": \"Lifespan in ${userLanguage}\" or null,\n \"breedingDifficultyEnum\": \"easy\" or \"moderate\" or \"difficult\" or \"very_difficult\" or null (EXACT English enum),\n \"breedingDifficulty\": \"Translated breeding difficulty in ${userLanguage}\" or null,\n \"breedingRequirements\": \"Breeding requirements in ${userLanguage}\" or null,\n \"sexualDifferences\": \"Sexual differences in ${userLanguage}\" or null,\n \"commonDiseases\": \"Common diseases in ${userLanguage}\" or null,\n \"diseaseSusceptibility\": \"Disease susceptibility in ${userLanguage}\" or null,\n \"specialCareConsiderations\": \"Special care in ${userLanguage}\" or null,\n \n \"confidence\": 0.0 to 1.0\n}\n\nIMPORTANT GUIDELINES:\n- Provide detailed, practical information\n- Consider safety heavily - prioritize toxicity/danger warnings\n- If you cannot identify clearly, set confidence below 0.5\n- If you can very confident, set confidence between 0.9 to 0.99 as you see fit\n- Use null for fields you cannot determine with confidence\n- Focus on observable details in the image${languageInstruction}${enumInstruction}\n\nIf the image contains no fish or is completely unclear, respond:\n{\n \"error\": \"No fish detected in image\",\n \"commonName\": null,\n \"scientificName\": null,\n \"alternativeNames\": null,\n \"family\": null,\n \"order\": null,\n \"averageLength\": null,\n \"averageWeight\": null,\n \"maximumLength\": null,\n \"maximumWeight\": null,\n \"distinctiveFeatures\": null,\n \"colorVariations\": null,\n \"sexualDimorphism\": null,\n \"nativeRange\": null,\n \"habitatType\": null,\n \"preferredDepth\": null,\n \"temperatureRange\": null,\n \"commonLocations\": null,\n \"edibilityStatusEnum\": null,\n \"edibilityStatus\": null,\n \"tasteDescription\": null,\n \"cookingMethods\": null,\n \"mercuryConcerns\": null,\n \"toxicityWarnings\": null,\n \"dangerousFeatures\": null,\n \"handlingPrecautions\": null,\n \"fishingMethods\": null,\n \"bestBaitLures\": null,\n \"bestSeasons\": null,\n \"bestTimes\": null,\n \"fightingCharacteristics\": null,\n \"isPopularSportFish\": false,\n \"conservationStatus\": null,\n \"isProtected\": false,\n \"isInvasive\": false,\n \"similarSpecies\": null,\n \"identificationDifferences\": null,\n \"isCommonPetFish\": false,\n \"minimumTankSize\": null,\n \"tankTypeEnum\": null,\n \"tankType\": null,\n \"waterTemperatureRange\": null,\n \"phRange\": null,\n \"waterHardness\": null,\n \"filtrationRequirements\": null,\n \"specialEquipment\": null,\n \"temperamentEnum\": null,\n \"temperament\": null,\n \"socialBehaviorEnum\": null,\n \"socialBehavior\": null,\n \"compatibleTankMates\": null,\n \"incompatibleTankMates\": null,\n \"isReefSafe\": false,\n \"difficultyLevelEnum\": null,\n \"difficultyLevel\": null,\n \"dietTypeEnum\": null,\n \"dietType\": null,\n \"foodRecommendations\": null,\n \"feedingFrequency\": null,\n \"lifespanInCaptivity\": null,\n \"breedingDifficultyEnum\": null,\n \"breedingDifficulty\": null,\n \"breedingRequirements\": null,\n \"sexualDifferences\": null,\n \"commonDiseases\": null,\n \"diseaseSusceptibility\": null,\n \"specialCareConsiderations\": null,\n \"confidence\": 0.0\n}`,\n },\n ];\n\n // Add all images to content array\n for (const base64Image of base64Images) {\n contentArray.push({\n type: \"image_url\",\n image_url: {\n url: `data:image/jpeg;base64,${base64Image}`,\n },\n });\n }\n\n const response = await openai.chat.completions.create({\n model: \"gpt-4o\",\n messages: [\n {\n role: \"user\",\n content: contentArray,\n },\n ],\n max_tokens: 2000,\n temperature: 0.3,\n });\n\n const content = response.choices[0].message.content;\n\n if (!content) {\n throw new Error(\"No response from OpenAI\");\n }\n\n // Clean the response\n let cleanedContent = content.trim();\n cleanedContent = cleanedContent.replace(/```json\\n?/g, \"\").replace(/```\\n?/g, \"\").trim();\n\n let result;\n try {\n result = JSON.parse(cleanedContent);\n console.log(\"✅ Parsed:\", result.commonName || result.error);\n } catch (parseError) {\n console.error(\"❌ Parse failed:\", cleanedContent);\n throw new Error(\"Invalid response format from AI\");\n }\n\n // Handle error responses\n if (result.error) {\n console.log(\"ℹ️ AI error:\", result.error);\n return new Response(\n JSON.stringify({\n success: false,\n error: result.error,\n }),\n {\n status: 200,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n // Validate required fields\n if (!result.commonName || !result.scientificName) {\n return new Response(\n JSON.stringify({\n success: false,\n error: \"We couldn't identify any fish. Please try with better photos.\",\n }),\n {\n status: 200,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n\n // Return successful identification with all fields\n return new Response(\n JSON.stringify({\n success: true,\n data: result,\n }),\n {\n status: 200,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n } catch (error) {\n console.error(\"❌ Error:\", error);\n\n let userMessage = \"Something went wrong. Please try again later.\";\n\n if (error instanceof Error) {\n if (\n error.message === \"No response from OpenAI\" ||\n error.message === \"Invalid response format from AI\"\n ) {\n userMessage = \"Something went wrong. Please try again later.\";\n } else {\n userMessage = error.message;\n }\n }\n\n return new Response(\n JSON.stringify({\n success: false,\n error: userMessage,\n }),\n {\n status: 500,\n headers: { ...corsHeaders, \"Content-Type\": \"application/json\" },\n },\n );\n }\n}","2026-02-10T10:16:09.938Z","FinID","https://www.val.town/x/mayank_yadav/FinID/code/","mayank_yadav",[285,288,290,293,296,298,301,304,307,310,313,316,318,321,323,325,326,328,330,332,335,338,341,343,346,349,351,353,356,359,362,365],{"_129":286,"_131":287,"_133":139},1,"import { OpenAI } from \"https://esm.sh/openai@4.20.1\";",{"_129":289,"_131":74,"_133":22},2,{"_129":291,"_131":292,"_133":22},3,"export default async function handler(req: Request): Promise {",{"_129":294,"_131":295,"_133":22},75," console.log(`📸 Processing ${base64Images.length} image(s)`);",{"_129":297,"_131":74,"_133":22},76,{"_129":299,"_131":300,"_133":139},77," const apiKey = Deno.env.get(\"OPENAI_API_KEY\");",{"_129":302,"_131":303,"_133":22},78," if (!apiKey) {",{"_129":305,"_131":306,"_133":139},79," console.error(\"❌ OPENAI_API_KEY not set\");",{"_129":308,"_131":309,"_133":22},80," return new Response(",{"_129":311,"_131":312,"_133":22},81," JSON.stringify({",{"_129":314,"_131":315,"_133":22},90," }",{"_129":317,"_131":74,"_133":22},91,{"_129":319,"_131":320,"_133":139},92," const openai = new OpenAI({ apiKey });",{"_129":322,"_131":74,"_133":22},93,{"_129":130,"_131":324,"_133":139}," console.log(\"🤖 Calling OpenAI for fish identification...\");",{"_129":135,"_131":74,"_133":22},{"_129":137,"_131":327,"_133":22}," // Language-specific instruction",{"_129":329,"_131":315,"_133":22},560,{"_129":331,"_131":74,"_133":22},561,{"_129":333,"_131":334,"_133":139},562," const response = await openai.chat.completions.create({",{"_129":336,"_131":337,"_133":22},563," model: \"gpt-4o\",",{"_129":339,"_131":340,"_133":22},564," messages: [",{"_129":342,"_131":74,"_133":22},575,{"_129":344,"_131":345,"_133":22},576," if (!content) {",{"_129":347,"_131":348,"_133":139},577," throw new Error(\"No response from OpenAI\");",{"_129":350,"_131":315,"_133":22},578,{"_129":352,"_131":74,"_133":22},579,{"_129":354,"_131":355,"_133":22},638," if (error instanceof Error) {",{"_129":357,"_131":358,"_133":22},639," if (",{"_129":360,"_131":361,"_133":139},640," error.message === \"No response from OpenAI\" ||",{"_129":363,"_131":364,"_133":22},641," error.message === \"Invalid response format from AI\"",{"_129":366,"_131":367,"_133":22},642," ) {",8,{"_106":370,"_108":371,"_110":372,"_112":373,"_114":374,"_116":375,"_118":376,"_120":377,"_122":378,"_124":-5,"_125":-5,"_126":379,"_160":273},"019a26dc-99bf-7269-a8f6-707670e63f81","https://www.val.town/x/lightweight/todoSweeper/code/README.md","1cb13b14-b360-11f0-bc6f-0224a6c84d84","README.md","# Notion Todo Sync\n\nAutomatically syncs Notion checkbox todos to a centralized database with intelligent owner resolution, relation mapping, and AI-powered summaries.\n\n## Table of Contents\n\n- [Quick Start](#quick-start)\n- [How It Works](#how-it-works)\n- [Property Mapping](#property-mapping)\n- [Owner Resolution](#owner-resolution)\n- [Other People Resolution](#other-people-resolution)\n- [Due Date Resolution](#due-date-resolution)\n- [Status Resolution](#status-resolution)\n- [Relation Mapping](#relation-mapping)\n- [Configuration Reference](#configuration-reference)\n- [Advanced Topics](#advanced-topics)\n\n---\n\n## Quick Start\n\n### Required Environment Variables\n\n| Variable | Description |\n| ---------------- | ------------------------------------------------------------------------------------- |\n| `NOTION_API_KEY` | Your Notion integration token ([get one here](https://www.notion.so/my-integrations)) |\n| `TODOS_DB_ID` | Database ID where todos are synced (32-char ID without hyphens) |\n\n### Recommended: OpenAI API Key\n\n| Variable | Description |\n| ---------------- | ----------------------------------------------------- |\n| `OPENAI_API_KEY` | OpenAI API key for AI features (strongly recommended) |\n\n**What AI does:**\n\n- Polishes raw todo text into clean, standalone summaries\n- Extracts owner from text patterns (e.g., \"Taylor to call Jordan\" → owner: Taylor)\n- Determines which date in the todo is the deadline\n- Extracts other people mentioned in text\n- Disambiguates when multiple contacts match a name\n\n**Without `OPENAI_API_KEY`:** Falls back to Val Town's shared OpenAI client (10 req/min limit). If AI is unavailable, the system still works but uses raw text, heading/@mention-only owner resolution, and default due dates.\n\n### Required Database Properties\n\nYour Todos database needs these properties at minimum:\n\n| Property Name | Type | Purpose |\n| --------------- | --------- | --------------------------------------- |\n| Name | Title | The todo text (cleaned up by AI) |\n| Source block ID | Rich text | Unique identifier to prevent duplicates |\n\n### Setup Steps\n\n1. Create a Notion integration and get the API key\n2. Make sure the integration has access to your databases\n3. Set the environment variables\n\n---\n\n## How It Works\n\n```\nNotion pages → Search for todos → Blob storage → Sync to database\n```\n\n1. **Search**: Scans recent pages for checkbox (`to_do`) blocks\n2. **Extract**: Captures text, @mentions, dates, page links, and surrounding context\n3. **Store**: Saves to blob storage with sync tracking\n4. **Sync**: Creates/updates pages in your Todos database with AI-powered summaries\n\n**What gets captured from a todo:**\n\n- Block text → Name (polished by AI)\n- @mentions → Owner and Other People\n- Dates in text → Due Date (AI interprets which is the deadline)\n- Page links → Relations (auto-mapped to matching databases)\n- Preceding heading → Owner context (if heading matches a contact name)\n- Checked/unchecked → Status (Done or Not started, if configured)\n\n**One-way sync with user override:**\n\nSync flows one direction: source pages → todos database. Updates from the source continue until a user edits the todo directly in the database. At that point, the database becomes authoritative for that todo—todoSweeper won't overwrite your explicit changes.\n\n---\n\n## Health Dashboard\n\nVisit the root URL of your val (`main.http.tsx`) to see a dashboard showing:\n\n- **Connection status**: Notion API and OpenAI connectivity\n- **Property mappings**: Which properties are configured and whether they exist in your database\n\nUse this to verify your setup is working and debug property mapping issues.\n\n---\n\n## Property Mapping\n\nThe system maps todo data to your database properties. Properties are matched by name.\n\n### Default Property Names\n\n| Key | Default Name | Required | Purpose |\n| ---------------------- | ----------------------- | -------- | -------------------------------- |\n| `name` | Name | **Yes** | Todo text (title property) |\n| `sourceBlockId` | Source block ID | **Yes** | Unique ID for deduplication |\n| `owner` | Owner | No | Person responsible for the todo |\n| `otherPeople` | Other people | No | Other people mentioned |\n| `dueDate` | Due date | No | When it's due |\n| `status` | _(disabled)_ | No | Checkbox sync (Done/Not started) |\n| `projects` | _(disabled)_ | No | Project relation |\n| `sourceBlockUrl` | Source block URL | No | Link to original block |\n| `sourcePageUrl` | Source page URL | No | Link to source page |\n| `sourceLastEditedTime` | Source last edited time | No | When source was edited |\n| `sourceAuthor` | Source author | No | Who created the block |\n| `sourceText` | Source text | No | Raw block text |\n| `links` | Links | No | URLs found in block |\n\n### Overriding Property Names\n\nIf your database uses different property names, override with environment variables:\n\n```\nTODOS_PROP_{key}=Your Property Name\n```\n\nThe `{key}` can be either format:\n\n- camelCase: `TODOS_PROP_otherPeople`\n- SCREAMING_SNAKE: `TODOS_PROP_OTHER_PEOPLE`\n\n| Key (camelCase) | Key (SCREAMING_SNAKE) |\n| ---------------------- | ------------------------- |\n| `name` | `NAME` |\n| `sourceBlockId` | `SOURCE_BLOCK_ID` |\n| `owner` | `OWNER` |\n| `otherPeople` | `OTHER_PEOPLE` |\n| `dueDate` | `DUE_DATE` |\n| `status` | `STATUS` |\n| `projects` | `PROJECTS` |\n| `sourceBlockUrl` | `SOURCE_BLOCK_URL` |\n| `sourcePageUrl` | `SOURCE_PAGE_URL` |\n| `sourceLastEditedTime` | `SOURCE_LAST_EDITED_TIME` |\n| `sourceAuthor` | `SOURCE_AUTHOR` |\n| `sourceText` | `SOURCE_TEXT` |\n| `links` | `LINKS` |\n\n**Examples (both formats work identically):**\n\n```bash\n# These two are equivalent:\nTODOS_PROP_dueDate=Deadline\nTODOS_PROP_DUE_DATE=Deadline\n\n# These two are equivalent:\nTODOS_PROP_otherPeople=Contacts\nTODOS_PROP_OTHER_PEOPLE=Contacts\n\n# These two are equivalent:\nTODOS_PROP_sourceBlockId=Block ID\nTODOS_PROP_SOURCE_BLOCK_ID=Block ID\n\n# Enable optional features (disabled by default):\nTODOS_PROP_status=Status # or TODOS_PROP_STATUS=Status\nTODOS_PROP_projects=Projects # or TODOS_PROP_PROJECTS=Projects\n```\n\n**To disable a property:** Set it to empty string:\n\n```bash\nTODOS_PROP_otherPeople= # or TODOS_PROP_OTHER_PEOPLE=\n```\n\n### Property Type Detection\n\nOwner and Other People can be either:\n\n- **People type**: Links to Notion workspace users\n- **Relation type**: Links to pages in another database (e.g., Contacts)\n\nThe system auto-detects the type from your database schema and adjusts matching behavior accordingly.\n\n---\n\n## Owner Resolution\n\nOwner is determined using a strict priority order. The first match wins.\n\n### Priority Order\n\n```\n┌─────────────────────────────────────────────────────────────┐\n│ Priority 1: HEADING MATCH │\n│ Does the preceding heading match someone in owner DB? │\n│ Example: \"### Alex\" matches \"Alex Johnson\" in Contacts │\n├─────────────────────────────────────────────────────────────┤\n│ MATCH → Use heading match (most reliable) │\n│ NO → Continue to Priority 2 │\n└─────────────────────────────────────────────────────────────┘\n ↓\n┌─────────────────────────────────────────────────────────────┐\n│ Priority 2: @MENTION │\n│ Is there an @mention in the todo? │\n│ Example: \"@Jane should review the PR\" │\n├─────────────────────────────────────────────────────────────┤\n│ YES → Use first @mention │\n│ NO → Continue to Priority 3 │\n└─────────────────────────────────────────────────────────────┘\n ↓\n┌─────────────────────────────────────────────────────────────┐\n│ Priority 3: AI EXTRACTION │\n│ AI extracts owner from todo text patterns │\n│ Example: \"Taylor to call Jordan\" → Taylor is owner │\n├─────────────────────────────────────────────────────────────┤\n│ Name extracted → Validate against owner database │\n│ FOUND → Use matched page/user │\n│ NOT FOUND → Owner = null (prevents hallucination) │\n└─────────────────────────────────────────────────────────────┘\n```\n\n### Heading Proximity Rules\n\nHeadings only apply to todos **directly beneath them**. Any intervening block resets the heading context.\n\n**Heading applies:**\n\n```\n### Alex\n- [ ] Todo 1 ← owner context: \"Alex\"\n- [ ] Todo 2 ← owner context: \"Alex\"\n```\n\n**Paragraph resets heading:**\n\n```\n### Alex\nSome intro text here\n- [ ] Todo 1 ← owner context: null (reset by paragraph)\n```\n\n**Empty line resets heading:**\n\n```\n### Alex\n\n- [ ] Todo 1 ← owner context: null (reset by empty line)\n```\n\nThis prevents distant headings from incorrectly matching todos far below them.\n\n### People vs Relation Type\n\n| Owner Type | Matches Against |\n| ---------- | ----------------------------- |\n| People | Notion workspace members only |\n| Relation | Any page in target database |\n\n**Recommendation:** Use Relation type pointing to a Contacts database. It's more flexible and works with both workspace members and external contacts.\n\n---\n\n## Other People Resolution\n\nOther People captures everyone mentioned in a todo except the owner.\n\n### Sources\n\n1. **@Mentions**: All @mentions (except the resolved owner) are added\n2. **AI-Extracted Names**: Names mentioned in text (only when Other People is a relation type)\n\n### AI-Extracted Name Matching\n\nWhen Other People is a relation property, AI extracts names from the todo text:\n\n```\n\"Call Kevin about the project\" → AI extracts \"Kevin\"\n```\n\n**Matching rules:**\n\n- Exact match first (case-insensitive)\n- Prefix match fallback: \"Kevin M\" matches \"Kevin McIntyre\"\n- Skipped if ambiguous (multiple matches)\n- Skipped if already the owner or an @mention\n\n---\n\n## Due Date Resolution\n\nAI determines the due date from your todo content.\n\n### How It Works\n\n1. AI examines all dates mentioned in the todo\n2. AI picks the one that looks like a deadline\n3. If no date in text, uses `DEFAULT_DUE_DATE` setting\n4. AI always returns a date (never leaves it blank)\n\n### Default Date Options\n\nSet `DEFAULT_DUE_DATE` environment variable:\n\n| Value | Meaning |\n| ------------------- | ------------------- |\n| `today` | Due today (default) |\n| `tomorrow` | Due tomorrow |\n| `one_week` | Due 7 days from now |\n| `end_of_week` | Due next Friday |\n| `next_business_day` | Due next weekday |\n\n---\n\n## Status Resolution\n\nWhen `TODOS_PROP_STATUS` is set, checkbox state syncs to your status property.\n\n| Checkbox State | Status Value |\n|----------------|--------------|\n| Checked | \"Done\" (or \"Complete\", \"Completed\", \"Finished\") |\n| Unchecked | First non-done option (e.g., \"Not started\", \"To Do\") |\n\nSupports Status, Select, and Checkbox property types. The system auto-detects your property type and finds matching options.\n\n---\n\n## Relation Mapping\n\nTodos are automatically linked to related pages (projects, tags, etc.) using these strategies.\n\n### Strategy 1: Explicit Matches (All Relations)\n\nWorks for any relation property.\n\n**1a: @Mentions**\nIf the todo @mentions a page that exists in a relation's target database, it's linked.\n\n```\n- [ ] Review [[Project Alpha]] specs\n→ Linked to \"Project Alpha\" in Projects\n```\n\n**1b: Heading Match**\nIf the preceding heading exactly matches a page name in a relation's target database, it's linked.\n\n```\n### Project Alpha\n- [ ] Review the specs\n→ Linked to \"Project Alpha\" in Projects\n```\n\n### Strategy 2: Contextual Matches (Projects Only)\n\nThese strategies only apply to the Projects relation. Other relations require explicit @mentions or heading matches.\n\n**2a: Source Page Relations**\nIf the source page has a relation property with matching values, those are inherited.\n\n```\nMeeting Notes page has Project = \"Acme Redesign\"\n→ All todos on that page inherit \"Acme Redesign\"\n```\n\n**2b: Source Page is Database Entry**\nIf the source page is a direct entry in the Projects database, todos link to that page.\n\n```\nTodo on \"Project Beta\" page (which is in Projects DB)\n→ Linked to \"Project Beta\"\n```\n\n**2c: Parent Page Traversal**\nFor nested pages, walks up the hierarchy (up to 3 levels) looking for 2a/2b matches.\n\n---\n\n## Configuration Reference\n\n### Required Variables\n\n| Variable | Description |\n| ---------------- | ---------------------------- |\n| `NOTION_API_KEY` | Notion integration token |\n| `TODOS_DB_ID` | Database ID for synced todos |\n\n### Property Overrides\n\n| Variable | Default | Description |\n| ------------------------------------ | ----------------------- | ------------------- |\n| `TODOS_PROP_NAME` | Name | Todo title property |\n| `TODOS_PROP_SOURCE_BLOCK_ID` | Source block ID | Deduplication ID |\n| `TODOS_PROP_OWNER` | Owner | Person responsible |\n| `TODOS_PROP_OTHER_PEOPLE` | Other people | Other mentions |\n| `TODOS_PROP_DUE_DATE` | Due date | Deadline |\n| `TODOS_PROP_STATUS` | _(disabled)_ | Checkbox sync |\n| `TODOS_PROP_PROJECTS` | _(disabled)_ | Project relation |\n| `TODOS_PROP_LINKS` | Links | URLs found |\n| `TODOS_PROP_SOURCE_BLOCK_URL` | Source block URL | Link to block |\n| `TODOS_PROP_SOURCE_PAGE_URL` | Source page URL | Link to page |\n| `TODOS_PROP_SOURCE_LAST_EDITED_TIME` | Source last edited time | Edit timestamp |\n| `TODOS_PROP_SOURCE_AUTHOR` | Source author | Block creator |\n| `TODOS_PROP_SOURCE_TEXT` | Source text | Raw text |\n\n### Behavior Settings\n\n| Variable | Default | Description |\n| ----------------------------- | ------- | ------------------------------- |\n| `DEFAULT_DUE_DATE` | today | Fallback when no date in todo |\n| `MIN_BLOCK_WORDS` | 5 | Skip todos with fewer words |\n| `BLOCK_STABILITY_MINUTES` | 0 | Wait before syncing (cron only) |\n| `RECENT_PAGES_LOOKBACK_HOURS` | 24 | How far back to search |\n| `CRONS_DISABLED` | false | Disable automatic sync |\n\n### Search Settings\n\n| Variable | Default | Description |\n| ------------------- | -------- | ------------------------------------- |\n| `SEARCH_BLOCK_TYPE` | to_do | Block type to search for |\n| `SEARCH_KEYWORDS` | _(none)_ | Additional keywords (comma-separated) |\n\n### Security & API Keys\n\n| Variable | Description |\n| ----------------------- | --------------------------------------------- |\n| `NOTION_WEBHOOK_SECRET` | API key for protecting endpoints |\n| `OPENAI_API_KEY` | For AI features (see Quick Start for details) |\n\n---\n\n## Advanced Topics\n\n### Setup vs Production Mode\n\nWhen first setting up, you may want to test on individual pages before letting the system sweep up all todos under your integration.\n\n**Setup mode** (`CRONS_DISABLED=true`):\n\n- Crons don't run - nothing syncs automatically\n- Use the `/tasks/todos/page` webhook to test on specific pages\n- Add a button in Notion that calls this endpoint to sync just that page\n- Validate property mappings, owner resolution, and relations work correctly\n\n**Production mode** (default, or `CRONS_DISABLED=false`):\n\n- Crons run every minute\n- All pages under your integration are scanned for todos\n- Todos automatically sync to your central database\n\nOnce you've validated everything works in setup mode, remove `CRONS_DISABLED` (or set to `false`) to enable automatic syncing.\n\n### Using Notion Buttons for Testing\n\nIn setup mode, you need `NOTION_WEBHOOK_SECRET` to authenticate webhook requests. This is **required for setup mode** but not needed for production mode (crons authenticate differently).\n\n**Step 1: Set the environment variable**\n\nSet `NOTION_WEBHOOK_SECRET` to any secure string:\n\n```\nNOTION_WEBHOOK_SECRET=your-secret-key-here\n```\n\n**Step 2: Get your val's URL**\n\nOpen `main.http.tsx` in Val Town. Copy the HTTP endpoint of that file (something like `https://.web.val.run`) - you'll need it for the button.\n\n**Step 3: Configure your Notion button**\n\nCreate a button in Notion with these settings:\n\n- **Action**: Send webhook\n- **URL**: Your val's base URL + `/tasks/todos/page` (e.g., `https://.web.val.run/tasks/todos/page`)\n- **Headers**: Add `X-API-KEY` with the exact same value as your `NOTION_WEBHOOK_SECRET`\n\nThe header value must match the environment variable exactly (case-sensitive). If they don't match, the request will be rejected with a 401 error.\n\n**What the endpoint does:**\n\n`POST /tasks/todos/page` syncs only the page containing the button, letting you validate property mappings, owner resolution, and relations before enabling crons for your entire workspace.\n\n### Cron Jobs\n\nThree crons run independently (when enabled):\n\n- **todoSearch.cron**: Scans recent pages for todos\n- **todoSync.cron**: Syncs blobs to database\n- **cacheWarm.cron**: Keeps health cache fresh\n\n### Sync Optimization\n\nThe system tracks sync state in blob storage:\n\n- **First sync**: Query + create = 2 API calls\n- **No changes**: Skip immediately = 0 API calls\n- **Block updated**: Direct update = 1 API call\n\nThis reduces API calls by 90%+ for stable todos.\n\n### Blob Storage\n\nBlob storage serves two purposes: **checkpointing** sync state and **caching** for performance.\n\n**Example blob** (one per todo block):\n\n```json\n{\n // Block identity\n \"block_id\": \"abc123-def456-...\",\n \"block_url\": \"https://notion.so/page#abc123\",\n \"page_url\": \"https://notion.so/page\",\n\n // Todo content\n \"todo_string\": \"Call Kevin about the project by Friday\",\n \"checked\": false,\n\n // Context for matching\n \"preceding_heading\": \"Alex\",\n \"people_mentions\": [{\"id\": \"...\", \"name\": \"Kevin\", \"email\": \"...\"}],\n \"date_mentions\": [\"2025-01-24\"],\n \"link_mentions\": [{\"text\": \"Project Alpha\", \"url\": \"...\", \"pageId\": \"...\"}],\n\n // Source page context (for project inheritance)\n \"source_page_database_id\": \"projects-db-id\",\n \"source_page_relations\": {\"Projects\": [\"page-id-1\"]},\n\n // Metadata\n \"last_edited_time\": \"2025-01-22T10:30:00.000Z\",\n \"author\": {\"id\": \"...\", \"name\": \"Jane\", \"email\": \"...\"},\n\n // Sync tracking\n \"sync_metadata\": {\n \"synced\": true,\n \"syncing\": false,\n \"target_page_id\": \"xyz789\",\n \"archived\": false\n }\n}\n```\n\n**Key fields explained:**\n\n| Field | Purpose |\n|-------|---------|\n| `sync_metadata.synced` | `false` = needs sync, `true` = already synced |\n| `sync_metadata.syncing` | Lock to prevent concurrent syncs (5-minute timeout) |\n| `sync_metadata.target_page_id` | Cached ID so updates don't require a database query |\n| `preceding_heading` | Used for owner matching when heading matches a contact name |\n| `source_page_relations` | Inherited project relations from the source page |\n\n**Checkpointing flow:**\n\n1. Search phase finds todo → saves blob with `synced: false`\n2. Save phase acquires lock (`syncing: true`)\n3. On success → marks `synced: true`, stores `target_page_id`\n4. On failure → releases lock (`syncing: false`), retries next cron run\n\n**Caching:**\n\n- Health data and relation mappings are cached separately with 60-second TTL\n- `cacheWarm.cron` keeps caches fresh for instant dashboard loads\n\n### Caching\n\nTwo caches with 1-minute TTL:\n\n- **Health cache**: Instant dashboard loads\n- **Relation cache**: Fast sync initialization\n\nProperty renames in Notion may take up to 1 minute to propagate.\n\n### Validation\n\nTodos are validated during search:\n\n- Must have at least `MIN_BLOCK_WORDS` words (default: 5)\n- Shorter blocks are skipped as not meaningful\n\n### Project Structure\n\n```\nbackend/\n├── controllers/ # Business logic\n├── services/ # External API integrations\n├── routes/ # HTTP handlers\n├── crons/ # Scheduled jobs\n└── utils/ # Helpers\nfrontend/ # React dashboard\nshared/ # Shared types\n```\n\n### Known Limitations\n\n**Notion Timestamp Precision**: Block edit times are rounded to the minute. If you edit a todo twice within 60 seconds, only the first edit syncs until a later edit. Checkbox toggles are explicitly compared to catch same-minute changes.\n","2026-02-09T19:14:27.840Z","todoSweeper","https://www.val.town/x/lightweight/todoSweeper/code/","lightweight",[380,383,385,388,390,393,396,399,401,404,407,409,412,414,417,419,420,422,424,425,428,431,434,436],{"_129":381,"_131":382,"_133":22},27,"| `TODOS_DB_ID` | Database ID where todos are synced (32-char ID without hyphens) |",{"_129":384,"_131":74,"_133":22},28,{"_129":386,"_131":387,"_133":139},29,"### Recommended: OpenAI API Key",{"_129":389,"_131":74,"_133":22},30,{"_129":391,"_131":392,"_133":22},31,"| Variable | Description |",{"_129":394,"_131":395,"_133":22},32,"| ---------------- | ----------------------------------------------------- |",{"_129":397,"_131":398,"_133":139},33,"| `OPENAI_API_KEY` | OpenAI API key for AI features (strongly recommended) |",{"_129":400,"_131":74,"_133":22},34,{"_129":402,"_131":403,"_133":22},35,"**What AI does:**",{"_129":405,"_131":406,"_133":22},41,"- Disambiguates when multiple contacts match a name",{"_129":408,"_131":74,"_133":22},42,{"_129":410,"_131":411,"_133":139},43,"**Without `OPENAI_API_KEY`:** Falls back to Val Town's shared OpenAI client (10 req/min limit). If AI is unavailable, the system still works but uses raw text, heading/@mention-only owner resolution, and default due dates.",{"_129":413,"_131":74,"_133":22},44,{"_129":415,"_131":416,"_133":22},45,"### Required Database Properties",{"_129":314,"_131":418,"_133":22},"Visit the root URL of your val (`main.http.tsx`) to see a dashboard showing:",{"_129":317,"_131":74,"_133":22},{"_129":319,"_131":421,"_133":139},"- **Connection status**: Notion API and OpenAI connectivity",{"_129":322,"_131":423,"_133":22},"- **Property mappings**: Which properties are configured and whether they exist in your database",{"_129":130,"_131":74,"_133":22},{"_129":426,"_131":427,"_133":22},425,"| ----------------------- | --------------------------------------------- |",{"_129":429,"_131":430,"_133":22},426,"| `NOTION_WEBHOOK_SECRET` | API key for protecting endpoints |",{"_129":432,"_131":433,"_133":139},427,"| `OPENAI_API_KEY` | For AI features (see Quick Start for details) |",{"_129":435,"_131":74,"_133":22},428,{"_129":437,"_131":438,"_133":22},429,"---","projectResults",[441,454],{"_106":442,"_108":443,"_112":444,"_122":445,"_124":446,"_125":447,"_448":449,"_116":450,"_451":452,"_453":289},"e28740fe-0450-11f1-a16b-42dde27851f2","https://www.val.town/x/kidjs/openai-agents/code/","openai-agents","kidjs","Template to use the OpenAI Agents SDK","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/4808a810-6001-47ed-dac6-b18afdf23100/square512","forked_branch_id","52cd3bfc-c630-11f0-a182-42dde27851f2","2026-02-07T18:14:46.820Z","user_id","019c3922-265b-77ad-8e98-81df6ce67c92","matchCount",{"_106":455,"_108":456,"_112":444,"_122":457,"_124":446,"_125":447,"_448":449,"_116":458,"_451":459,"_453":289},"d8b46b26-025a-11f1-a1ec-42dde27851f2","https://www.val.town/x/EatPraySin/openai-agents/code/","EatPraySin","2026-02-05T06:21:03.087Z","019c224d-4d79-75c7-85ed-0dd952082964","userResults",[],"docResults",[],"totalCount",107,"paginationCount","totalFileResults",105,"totalProjectResults","totalUserResults","totalDocResults","executionTimeMs",303,"results",{"_103":104,"_439":440,"_460":461,"_462":463,"_467":468,"_469":289,"_470":33,"_471":33,"_96":286,"_476":477,"_478":286,"_479":102,"_472":473},"pageSize",10,"totalPages","resultType","actionData","errors"]

Search

Vals2

Vals
2