[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_176":177,"_178":179},"routes/_app.x.$handle.$name",{"_11":12,"_13":14,"_15":-7,"_18":19,"_17":-7,"_133":-5,"_20":134,"_22":135,"_136":137,"_138":139,"_26":140,"_24":141,"_142":-5,"_143":-5,"_144":51,"_145":146,"_147":43,"_148":-5,"_149":-5,"_150":151},"routes/_app.x.$handle.$name.code.$",{"_11":12,"_13":14,"_15":16,"_17":-7,"_18":19,"_20":21,"_22":23,"_24":25,"_26":27,"_28":-5,"_29":-5,"_30":-5,"_31":-5,"_32":-5},"handle","neverstew","name","aimemory","branchName","main","version","path","index.html","project",{"_33":80,"_13":14,"_44":81,"_82":-5,"_83":51,"_84":85,"_86":85,"_87":-5,"_88":-5,"_89":90,"_91":59,"_92":43,"_93":43,"_94":-5,"_95":76,"_96":97,"_98":-5,"_99":-5,"_100":-5,"_101":-5,"_102":103,"_104":103,"_105":106,"_107":106,"_108":106,"_109":103,"_110":103,"_111":112,"_113":114,"_115":116,"_117":51},"tree",[54,55,56],"file",{"_33":34,"_13":19,"_35":36,"_37":-5,"_38":39,"_40":41,"_42":43,"_44":45,"_17":46,"_47":24,"_48":49,"_50":51,"_52":-5,"_18":53},"contents",[],"upstream","pull","ahead","behind","envVars","id","b472eb22-31a2-11f0-a898-569c3dd06744","fileId","8010d03a-31a0-11f0-a3de-569c3dd06744","parentId","branchId","36cf1ef4-31a0-11f0-a4fa-569c3dd06744","branchVersion",20,"latest",true,"createdAt",["D",1747323533025],5,"fileType","content","\n\n\n \n \n AI Agent with Memory\n \n \n \n\n\n

AI Agent with Memory

\n \n

Ask a Question

\n \n

Response:

Ask a question to see a response here.

\n \n

Relevant Memories:

\n \n

Save New Memory

\n \n

\n\n \n\n","deleted",false,"label",[19],{"_33":71,"_13":72,"_35":73,"_37":-5,"_38":39,"_40":74,"_42":43,"_44":75,"_17":76,"_47":24,"_48":77,"_50":51,"_52":-5,"_65":78,"_18":79},{"_33":34,"_13":19,"_35":36,"_37":-5,"_38":39,"_40":41,"_42":43,"_44":68,"_17":46,"_47":24,"_48":49,"_50":51,"_52":-5,"_65":69,"_18":70},{"_33":57,"_13":58,"_35":59,"_37":-5,"_38":39,"_40":60,"_42":43,"_44":61,"_17":62,"_47":63,"_48":64,"_50":51,"_52":-5,"_65":66,"_18":67},"997d01a4-31a2-11f0-83f9-569c3dd06744","index.ts","6fae9d08-31a0-11f0-8def-569c3dd06744",18,["D",1747323487796],12,"http","import { sqlite } from \"https://esm.town/v/stevekrouse/sqlite\";\nimport { OpenAI } from \"https://esm.town/v/std/openai\";\nimport { readFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n\n// Initialize OpenAI client\nconst openai = new OpenAI();\n\n// Database setup\nconst MEMORY_TABLE = \"ai_agent_memory\";\n\n// Initialize the database table\nasync function initializeDatabase() {\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS ${MEMORY_TABLE} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n content TEXT NOT NULL,\n tags TEXT,\n created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP\n )\n `);\n}\n\n// Save a new memory snippet\nasync function saveMemory(content: string, tags: string = \"\") {\n await sqlite.execute(\n `INSERT INTO ${MEMORY_TABLE} (content, tags) VALUES (?, ?)`,\n [content, tags]\n );\n return { success: true, message: \"Memory saved successfully\" };\n}\n\n// Search for relevant memories based on a query\nasync function searchMemories(query: string): Promise> {\n // Simple search implementation - could be improved with embeddings or more sophisticated search\n const results = await sqlite.execute(\n `SELECT * FROM ${MEMORY_TABLE} WHERE content LIKE ? OR tags LIKE ?`,\n [`%${query}%`, `%${query}%`]\n );\n \n return results.rows as Array<{ id: number; content: string; tags: string; created_at: string }>;\n}\n\n// Get all memories from the database\nasync function getAllMemories(): Promise> {\n const results = await sqlite.execute(\n `SELECT * FROM ${MEMORY_TABLE} ORDER BY created_at DESC`\n );\n \n return results.rows as Array<{ id: number; content: string; tags: string; created_at: string }>;\n}\n\n// Fallback response generator when OpenAI is not available\nfunction generateFallbackResponse(query: string, memories: Array<{ content: string; tags?: string }> = []) {\n if (memories.length === 0) {\n return `I don't have any memories stored yet. You can teach me by saving some information.`;\n }\n \n // Find potentially relevant memories by simple text matching\n const relevantMemories = memories.filter(m => \n m.content.toLowerCase().includes(query.toLowerCase()) || \n (m.tags && m.tags.toLowerCase().includes(query.toLowerCase()))\n );\n \n if (relevantMemories.length === 0) {\n return `I have ${memories.length} memories stored, but none seem directly relevant to \"${query}\". You can ask about something else or teach me about this topic by saving a new memory.`;\n }\n \n return `Based on my ${memories.length} stored memories, I found ${relevantMemories.length} that might be relevant to \"${query}\":\\n\\n` +\n relevantMemories.map((m, i) => `${i+1}. ${m.content}`).join('\\n\\n') +\n '\\n\\nIs there anything specific about this information you would like to know?';\n}\n\n// Generate a response using OpenAI, incorporating memories if available\nasync function generateResponse(query: string, memories: Array<{ content: string; tags?: string }> = []) {\n try {\n // Format all memories as context\n const memoryContext = memories.length > 0 \n ? `I have the following memories in my database (${memories.length} total):\\n\\n${\n memories.map((m, i) => {\n const tags = m.tags ? ` [Tags: ${m.tags}]` : '';\n return `Memory ${i+1}:${tags}\\n${m.content}`;\n }).join('\\n\\n')\n }\\n\\n`\n : \"I don't have any memories stored yet.\\n\\n\";\n \n console.log(`Sending to OpenAI with ${memories.length} memories in context`);\n \n const response = await openai.chat.completions.create({\n model: \"gpt-4o-mini\",\n messages: [\n { \n role: \"system\", \n content: \"You are an AI assistant with memory. You have access to all your stored memories and should use them to provide accurate and helpful responses. If the question relates to information in your memories, reference that information directly.\" \n },\n { \n role: \"user\", \n content: `${memoryContext}Based on ALL the memories above, please respond to this question: ${query}` \n }\n ],\n max_tokens: 500,\n });\n \n if (!response.choices || response.choices.length === 0) {\n console.error('OpenAI returned no choices:', response);\n return generateFallbackResponse(query, memories);\n }\n \n return response.choices[0].message.content || generateFallbackResponse(query, memories);\n } catch (error) {\n console.error('Error generating response with OpenAI:', error);\n return generateFallbackResponse(query, memories);\n }\n}\n\n// Process the incoming request\nasync function processRequest(req: Request) {\n // Initialize database on each request (only creates if not exists)\n await initializeDatabase();\n \n try {\n const { action, query, content, tags } = await req.json();\n console.log('Received request:', { action, query, content, tags });\n \n // Handle different actions\n switch (action) {\n case \"query\":\n try {\n // Get all memories instead of just searching for relevant ones\n const allMemories = await getAllMemories();\n console.log('All memories:', allMemories);\n \n // Also get relevant memories for highlighting in the UI\n const relevantMemories = await searchMemories(query);\n console.log('Relevant memories:', relevantMemories);\n \n // Generate response using ALL memories\n const response = await generateResponse(query, allMemories);\n console.log('Generated response:', response);\n \n return new Response(JSON.stringify({ \n response,\n memories_used: allMemories.length,\n // Return relevant memories for highlighting in the UI\n memories: relevantMemories,\n all_memories_count: allMemories.length\n }), {\n headers: { \"Content-Type\": \"application/json\" }\n });\n } catch (error) {\n console.error('Error in query processing:', error);\n return new Response(JSON.stringify({ \n error: \"Failed to process your query. Please try again.\",\n details: error.message\n }), {\n status: 500,\n headers: { \"Content-Type\": \"application/json\" }\n });\n }\n \n case \"save\":\n // Save new memory\n const result = await saveMemory(content, tags);\n return new Response(JSON.stringify(result), {\n headers: { \"Content-Type\": \"application/json\" }\n });\n \n default:\n return new Response(JSON.stringify({ \n error: \"Invalid action. Use 'query' to ask questions or 'save' to store memories.\" \n }), {\n status: 400,\n headers: { \"Content-Type\": \"application/json\" }\n });\n }\n } catch (error) {\n console.error(\"Error processing request:\", error);\n return new Response(JSON.stringify({ error: \"Failed to process request\" }), {\n status: 500,\n headers: { \"Content-Type\": \"application/json\" }\n });\n }\n}\n\n// HTTP handler\nexport default async function(req: Request) {\n // Handle OPTIONS for CORS\n if (req.method === \"OPTIONS\") {\n return new Response(null, {\n headers: {\n \"Access-Control-Allow-Origin\": \"*\",\n \"Access-Control-Allow-Methods\": \"POST, GET, OPTIONS\",\n \"Access-Control-Allow-Headers\": \"Content-Type\"\n }\n });\n }\n \n // Handle GET requests - serve the HTML page\n if (req.method === \"GET\") {\n try {\n const html = await readFile(\"/index.html\", import.meta.url);\n return new Response(html, {\n headers: {\n \"Content-Type\": \"text/html\",\n \"Access-Control-Allow-Origin\": \"*\"\n }\n });\n } catch (error) {\n console.error(\"Error serving HTML:\", error);\n return new Response(\"Error loading the application\", {\n status: 500,\n headers: { \"Content-Type\": \"text/plain\" }\n });\n }\n }\n \n // Handle POST requests for API operations\n if (req.method === \"POST\") {\n // Process the request\n const response = await processRequest(req);\n \n // Add CORS headers to the response\n const headers = new Headers(response.headers);\n headers.set(\"Access-Control-Allow-Origin\", \"*\");\n \n return new Response(response.body, {\n status: response.status,\n statusText: response.statusText,\n headers\n });\n }\n \n // If not GET, POST or OPTIONS, return method not allowed\n return new Response(JSON.stringify({ error: \"Method not allowed\" }), {\n status: 405,\n headers: { \n \"Content-Type\": \"application/json\",\n \"Access-Control-Allow-Origin\": \"*\"\n }\n });\n}","children",[],[58],["D",1747323533025],[],[19],"8ab1f30c-31a0-11f0-8807-569c3dd06744","README.md","8ab1f3d4-31a0-11f0-8807-569c3dd06744",3,["D",1747322603987],0,"# AI Agent with Memory\n\nThis is an AI agent that can save information to its memory and retrieve it when answering questions. The agent uses SQLite for persistent storage and OpenAI's API for generating responses.\n\n## Features\n\n- **Memory Storage**: Save snippets of information with optional tags\n- **Memory Retrieval**: When asked questions, the agent searches its memory for relevant information\n- **Context-Aware Responses**: Incorporates found memories into its responses\n- **Simple Web Interface**: Easy-to-use UI for interacting with the agent\n\n## How It Works\n\n1. **Database**: Uses SQLite to store memory snippets with content, tags, and timestamps\n2. **Memory Search**: Performs simple text matching to find relevant memories\n3. **Response Generation**: Uses OpenAI to generate responses that incorporate found memories\n4. **Web Interface**: Provides a clean UI for asking questions and saving new memories\n\n## Usage\n\n### Asking Questions\n\n1. Enter your question in the \"Ask a Question\" section\n2. Click \"Ask Question\"\n3. The agent will search its memory for relevant information and incorporate it into the response\n4. If memories were used, they will be displayed below the response\n\n### Saving Memories\n\n1. Enter the information you want the agent to remember in the \"Save New Memory\" section\n2. Optionally add tags (comma-separated) to help with retrieval\n3. Click \"Save Memory\"\n\n## Technical Details\n\n- **Backend**: TypeScript with Deno runtime\n- **Database**: SQLite for persistent storage\n- **AI**: OpenAI API for generating responses\n- **Frontend**: HTML, JavaScript, and Twind (Tailwind CSS in JS)\n\n## Future Improvements\n\n- Implement vector embeddings for more accurate memory retrieval\n- Add memory management features (edit, delete, categorize)\n- Improve search with semantic similarity rather than just text matching\n- Add authentication to create personal memory agents for different users",[],[72],"36ce875a-31a0-11f0-a4fa-569c3dd06744",["D",1747322463249],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","imageUrl","userId","8acb974a-4b91-4d4a-b000-6f28827b7e16","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount","orgId","b74060f0-0f1d-44f1-bee2-bdb9c9cf72e4","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_11":12,"_132":128},"org",{"_11":12,"_126":-5,"_127":128,"_89":90,"_129":130,"_131":51},"branch",{"_17":41,"_13":16,"_33":39,"_118":80,"_119":-5,"_120":-5,"_121":-5,"_122":-5,"_123":-5,"_124":14,"_125":12},"isOwned","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","displayName","avatar","https://secure.gravatar.com/avatar/109fa713caf4e083c57622b7c6dad365.jpg?s=200&d=identicon","tier","free","canRestrictVals","profileImageUrl","adoptOrgId",{"_33":80,"_13":14,"_44":172,"_82":-5,"_83":51,"_84":85,"_86":85,"_87":-5,"_88":-5,"_89":90,"_91":59,"_92":43,"_93":43,"_94":-5,"_95":76,"_96":97,"_98":-5,"_99":-5,"_100":-5,"_101":-5,"_102":103,"_104":103,"_105":106,"_107":106,"_108":106,"_109":103,"_110":103,"_111":173,"_113":174,"_115":175,"_117":51},[163,164,165],"files",[163,164,165],"counts",{"_153":76,"_154":155,"_156":76,"_32":157,"_158":159},[],{"_33":34,"_13":19,"_35":36,"_37":-5,"_38":39,"_40":41,"_42":43,"_44":45,"_17":46,"_47":24,"_48":49,"_50":51,"_52":-5,"_18":152},"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},[19],"pulls","branches",1,"forks",{"_160":76,"_162":76},"connections",{"_160":76,"_161":76},"count","unfilledCount","emptyCount",{"_33":71,"_13":72,"_35":73,"_37":-5,"_38":39,"_40":74,"_42":43,"_44":75,"_17":76,"_47":24,"_48":77,"_50":51,"_52":-5,"_65":170,"_18":171},{"_33":34,"_13":19,"_35":36,"_37":-5,"_38":39,"_40":41,"_42":43,"_44":68,"_17":46,"_47":24,"_48":49,"_50":51,"_52":-5,"_65":168,"_18":169},{"_33":57,"_13":58,"_35":59,"_37":-5,"_38":39,"_40":60,"_42":43,"_44":61,"_17":62,"_47":63,"_48":64,"_50":51,"_52":-5,"_65":166,"_18":167},[],[58],[],[19],[],[72],["D",1747322463249],{"_11":12,"_132":128},{"_11":12,"_126":-5,"_127":128,"_89":90,"_129":130,"_131":51},{"_17":41,"_13":16,"_33":39,"_118":80,"_119":-5,"_120":-5,"_121":-5,"_122":-5,"_123":-5,"_124":14,"_125":12},"ctx",{"_111":-5,"_149":209,"_210":211,"_212":213,"_214":51,"_215":51,"_216":51,"_217":43,"_218":-5,"_219":-7,"_220":-7,"_221":76,"_222":51,"_223":51,"_224":-5,"_225":51,"_226":51,"_227":51,"_228":-5},"ENV",{"_180":181,"_182":183,"_184":16,"_185":186,"_187":188,"_189":190,"_191":192,"_193":194,"_195":196,"_197":198,"_199":200,"_201":202,"_203":204,"_205":206,"_207":208},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]