openmates 0.15.0-alpha.44 → 0.15.0-alpha.45

package/dist/{chunk-LAAY2YYX.js → chunk-F2G7QKBM.js}

@@ -1284,6 +1284,7 @@ function resolveServerPath(flags) {
 import { createRequire } from "module";
 var require2 = createRequire(import.meta.url);
 var WebSocket = require2("ws");
+var DEFAULT_ASYNC_EMBED_WAIT_MS = 3e5;
 var SUB_CHAT_EVENT_TYPES = /* @__PURE__ */ new Set([
   "spawn_sub_chats",
   "sub_chat_progress",
@@ -1667,7 +1668,7 @@ var OpenMatesWsClient = class {
   collectAiResponse(userMessageId, chatId, options) {
     const timeoutMs = options?.timeoutMs ?? 9e4;
     const onStream = options?.onStream;
-    const asyncEmbedWaitMs = options?.asyncEmbedWaitMs ?? 12e4;
+    const asyncEmbedWaitMs = options?.asyncEmbedWaitMs ?? DEFAULT_ASYNC_EMBED_WAIT_MS;
     return new Promise((resolve8, reject) => {
       let latestContent = "";
       let messageId = null;
@@ -8771,6 +8772,14 @@ var OpenMatesClient = class _OpenMatesClient {
             category = resp.category;
             modelName = resp.modelName;
             assistantMessageId = assistantId;
+            await this.persistStreamedEmbeds({
+              ws,
+              embeds: resp.embeds,
+              chatId,
+              chatKeyBytes,
+              fallbackMessageId: assistantId,
+              ownerId
+            });
             clearSyncCache(teamId);
             await persistCompressionCheckpoints(resp.compressionCheckpoints);
             const mateName2 = category ? MATE_NAMES[category] ?? null : null;
@@ -40665,6 +40674,56 @@ embed_ref: result-XhI`,
   }
 };
 
+// ../ui/src/demo_chats/data/example_chats/screenshot-to-html-pricing-card.ts
+var screenshotToHtmlPricingCardChat = {
+  chat_id: "example-screenshot-to-html-pricing",
+  slug: "screenshot-to-html-pricing-card",
+  title: "example_chats.screenshot_to_html_pricing_card.title",
+  summary: "example_chats.screenshot_to_html_pricing_card.summary",
+  icon: "code",
+  category: "design",
+  keywords: ["screenshot to HTML", "HTML card", "visual AI", "frontend code", "design to code"],
+  follow_up_suggestions: ["example_chats.screenshot_to_html_pricing_card.follow_up_1", "example_chats.screenshot_to_html_pricing_card.follow_up_2", "example_chats.screenshot_to_html_pricing_card.follow_up_3", "example_chats.screenshot_to_html_pricing_card.follow_up_4", "example_chats.screenshot_to_html_pricing_card.follow_up_5", "example_chats.screenshot_to_html_pricing_card.follow_up_6"],
+  messages: [
+    {
+      "id": "6eec58cb-3d75-4726-9ae7-70549f6e8bc5",
+      "role": "user",
+      "content": "example_chats.screenshot_to_html_pricing_card.message_1",
+      "created_at": 1784765315
+    },
+    {
+      "id": "a8863386-c6d7-5c48-a7a7-36cc5bfbc5d0",
+      "role": "assistant",
+      "content": "example_chats.screenshot_to_html_pricing_card.message_2",
+      "created_at": 1784765522,
+      "user_message_id": "6eec58cb-3d75-4726-9ae7-70549f6e8bc5",
+      "response_credits": 250,
+      "category": "design",
+      "model_name": "Gemini 3.5 Flash-Lite"
+    }
+  ],
+  embeds: [
+    {
+      "embed_id": "395f8816-7e0b-4753-8058-f1f5c49d0482",
+      "type": "code",
+      "content": 'type: code\napp_id: code\nskill_id: image_to_html\nlanguage: html\ncode: "<!DOCTYPE html>\\n<html lang=\\"en\\">\\n<head>\\n  <meta charset=\\"UTF-8\\">\\n  <meta name=\\"viewport\\" content=\\"width=1440, height=1200, initial-scale=1.0\\">\\n  <title>Pricing Card</title>\\n  <style>\\n    * {\\n      box-sizing: border-box;\\n      margin: 0;\\n      padding: 0;\\n    }\\n\\n    body {\\n      width: 1440px;\\n      height: 1200px;\\n      margin: 0;\\n      padding: 0;\\n      display: flex;\\n      align-items: center;\\n      justify-content: center;\\n      background-color: #f4f6fc;\\n      background-image: \\n        radial-gradient(circle at 100% 100%, #fff4eb 0%, rgba(255, 244, 235, 0) 55%),\\n        radial-gradient(circle at 0% 0%, #e9effd 0%, rgba(233, 239, 253, 0) 55%);\\n      font-family: -apple-system, BlinkMacSystemFont, \\"Segoe UI\\", Roboto, Helvetica, Arial, sans-serif;\\n      -webkit-font-smoothing: antialiased;\\n      -moz-osx-font-smoothing: grayscale;\\n      overflow: hidden;\\n    }\\n\\n    .card {\\n      width: 418px;\\n      background: #ffffff;\\n      border-radius: 28px;\\n      padding: 42px 40px;\\n      box-shadow: \\n        0 32px 64px -12px rgba(110, 120, 175, 0.14),\\n        0 12px 28px -8px rgba(0, 0, 0, 0.02);\\n      display: flex;\\n      flex-direction: column;\\n    }\\n\\n    .badge {\\n      display: inline-flex;\\n      align-items: center;\\n      gap: 7px;\\n      background-color: #e8effd;\\n      color: #3b5bfd;\\n      font-size: 12.5px;\\n      font-weight: 600;\\n      padding: 5px 12px 5px 10px;\\n      border-radius: 100px;\\n      width: fit-content;\\n      margin-bottom: 24px;\\n    }\\n\\n    .badge-dot {\\n      width: 6px;\\n      height: 6px;\\n      background-color: #3b5bfd;\\n      border-radius: 50%;\\n    }\\n\\n    .title {\\n      font-size: 36px;\\n      font-weight: 800;\\n      color: #151828;\\n      line-height: 1.14;\\n      letter-spacing: -0.03em;\\n      margin-bottom: 14px;\\n    }\\n\\n    .description {\\n      font-size: 14.5px;\\n      line-height: 1.48;\\n      color: #676f83;\\n      margin-bottom: 32px;\\n    }\\n\\n    .price-container {\\n      display: flex;\\n      align-items: baseline;\\n      gap: 8px;\\n      margin-bottom: 32px;\\n    }\\n\\n    .price {\\n      font-size: 48px;\\n      font-weight: 800;\\n      color: #151828;\\n      line-height: 1;\\n      letter-spacing: -0.035em;\\n    }\\n\\n    .price-period {\\n      font-size: 14px;\\n      font-weight: 500;\\n      color: #676f83;\\n    }\\n\\n    .features {\\n      display: flex;\\n      flex-direction: column;\\n      gap: 16px;\\n      margin-bottom: 36px;\\n    }\\n\\n    .feature-item {\\n      display: flex;\\n      align-items: center;\\n      gap: 12px;\\n    }\\n\\n    .feature-icon {\\n      width: 20px;\\n      height: 20px;\\n      border-radius: 50%;\\n      background-color: #d8f7e5;\\n      display: flex;\\n      align-items: center;\\n      justify-content: center;\\n      flex-shrink: 0;\\n    }\\n\\n    .feature-icon-dot {\\n      width: 6px;\\n      height: 6px;\\n      border-radius: 50%;\\n      background-color: #02b153;\\n    }\\n\\n    .feature-text {\\n      font-size: 14.5px;\\n      font-weight: 700;\\n      color: #151828;\\n      letter-spacing: -0.01em;\\n    }\\n\\n    .btn {\\n      width: 100%;\\n      height: 52px;\\n      background: linear-gradient(90deg, #405bf8 0%, #813cfc 100%);\\n      color: #ffffff;\\n      border: none;\\n      border-radius: 16px;\\n      font-size: 15px;\\n      font-weight: 600;\\n      cursor: pointer;\\n      box-shadow: 0 10px 24px -2px rgba(105, 70, 255, 0.42);\\n      display: flex;\\n      align-items: center;\\n      justify-content: center;\\n      font-family: inherit;\\n    }\\n  </style>\\n</head>\\n<body>\\n  <div class=\\"card\\">\\n    <div class=\\"badge\\">\\n      <span class=\\"badge-dot\\"></span>\\n      Team Plan\\n    </div>\\n    \\n    <h1 class=\\"title\\">Design faster with<br>visual AI</h1>\\n    \\n    <p class=\\"description\\">\\n      Turn screenshots into clean front-end code and<br>review the result in seconds.\\n    </p>\\n    \\n    <div class=\\"price-container\\">\\n      <span class=\\"price\\">$29</span>\\n      <span class=\\"price-period\\">/ seat monthly</span>\\n    </div>\\n    \\n    <div class=\\"features\\">\\n      <div class=\\"feature-item\\">\\n        <div class=\\"feature-icon\\">\\n          <div class=\\"feature-icon-dot\\"></div>\\n        </div>\\n        <span class=\\"feature-text\\">Unlimited design imports</span>\\n      </div>\\n      <div class=\\"feature-item\\">\\n        <div class=\\"feature-icon\\">\\n          <div class=\\"feature-icon-dot\\"></div>\\n        </div>\\n        <span class=\\"feature-text\\">Responsive HTML export</span>\\n      </div>\\n      <div class=\\"feature-item\\">\\n        <div class=\\"feature-icon\\">\\n          <div class=\\"feature-icon-dot\\"></div>\\n        </div>\\n        <span class=\\"feature-text\\">Visual QA reports</span>\\n      </div>\\n    </div>\\n    \\n    <button class=\\"btn\\">Start building</button>\\n  </div>\\n</body>\\n</html>"\nfilename: index.html\nembed_ref: code-395f88\nstatus: finished\nline_count: 200\nimage_to_html_metadata:\n  correction_passes_used: 2\n  validation_warnings[0]:\n  usage:\n    model: gemini-3.6-flash\n    input_tokens: 19416\n    output_tokens: 5272\n    cache_read_tokens: 0\n    cache_write_tokens: 0\n    e2b_render_seconds: 100.38043770287186\n    correction_passes_used: 2\n    validation_repair_attempts: 0\n    extracted_asset_count: 0\n    visual_diff_score: 0.0149640522875817\n    duration_second: 100.38043770287186\n    provider_cost_usd: 0.068664\n    provider_cost_credits: 69\n    model_credits: 215\n    e2b_credits: 10\n    e2b_started_minutes: 2\n    reserved_credits: 1500\n    credits_charged: 225\n    credits_refunded: 1275',
+      "parent_embed_id": null,
+      "embed_ids": null
+    },
+    {
+      "embed_id": "fcfe81c4-5baa-4580-be67-d5bec235d2f4",
+      "type": "image",
+      "content": "type: image\napp_id: images\nskill_id: upload\nstatus: finished\nfilename: simple-card-input.png\nembed_ref: simple-card-input-png-fcfe81c4-5baa-4580-afa07b\ncontent_hash: efb02bdf09ab6665c910dc9b94ab00384280148bf633d9d9d0c6ce758a253ed4\nai_detection:\n  ai_generated: 0.99\n  provider: sightengine\n  status: success\n  error: null",
+      "parent_embed_id": null,
+      "embed_ids": null
+    }
+  ],
+  metadata: {
+    featured: true,
+    order: 118
+  }
+};
+
 // ../ui/src/demo_chats/exampleChatData.ts
 var ALL_EXAMPLE_CHATS = [
   giganticAirplanesChat,
@@ -40764,7 +40823,8 @@ var ALL_EXAMPLE_CHATS = [
   dashboardSidebarSvgIconsChat,
   vitalFarmsSecFinancialsChat,
   financeCashFlowOverviewChat,
-  privacyFirstLocalAiWorkspaceChat
+  privacyFirstLocalAiWorkspaceChat,
+  screenshotToHtmlPricingCardChat
 ].sort((a, b) => a.metadata.order - b.metadata.order);
 
 // ../ui/src/i18n/locales/en.json
@@ -50133,6 +50193,38 @@ Would you like to start executing on any specific pillar of this plan\u2014such
         text: '```json\n{"type": "app_skill_use", "embed_id": "2f72870d-27c9-4e6a-82a0-ac6cef2b8839", "app_id": "code", "skill_id": "search_repos", "query": "vector database rust"}\n```\n\n```json\n{"type": "app_skill_use", "embed_id": "bb757fef-77ab-4050-b3be-a62d0c177b72", "app_id": "code", "skill_id": "search_repos", "query": "vector search rust"}\n```\n\n```json\n{"type": "app_skill_use", "embed_id": "5358df62-b716-487b-a554-36c8c0ad0cf9", "app_id": "code", "skill_id": "search_repos", "query": "qdrant/qdrant"}\n```\n\n```json\n{"type": "app_skill_use", "embed_id": "70951676-9c32-4eac-bfe7-97e7b3888ce1", "app_id": "code", "skill_id": "search_repos", "query": "vector database language:rust"}\n```\n\nRust has become a premier language for building vector databases due to its zero-cost abstractions, memory safety, and C++ like performance\u2014qualities that are essential for handling millions of high-dimensional embeddings.\n\nBased on active, open-source repositories on GitHub, here is a comparison of the top Rust-based vector database engines and extensions available for self-hosting.\n\n### \u{1F4CA} Comparison Overview\n\n```json\n{"type": "sheet", "embed_id": "3895dddf-e6d3-4048-aabb-6687f38c788d"}\n```\n\n\n*\\*Activity is based on the most recent code commits as of June 2026.*\n\n---\n\n### \u{1F50D} Detailed Breakdown\n\n#### 1. The Industry Standard: Qdrant\n[Qdrant](embed:qdrant-B6U) is the undisputed leader when it comes to pure-Rust vector databases. It was built from the ground up for high-performance vector similarity search.\n*   **Pros:** Production-ready, massive community support, supports complex filtering payloads alongside vector search, cloud-native architecture, and includes a built-in Web UI.\n*   **Use Case:** Ideal if you are building production-level AI applications, recommendation systems, or semantic search architectures and want a standalone, horizontally scalable vector engine.\n\n#### 2. The Rising Graph-Vector Hybrid: HelixDB\n[HelixDB](embed:helix-db-SlF) is a relatively new but heavily trending database that combines graph database capabilities with vector embeddings.\n*   **Pros:** By combining graph nodes with vector embeddings, it\'s highly optimized for Retrieval-Augmented Generation (RAG) tasks where you need both semantic similarity (vectors) and explicit relationship context (graphs).\n*   **Use Case:** Excellent for advanced Agentic AI architectures, RAG, and knowledge graphs where standard vector search lacks relational context.\n\n#### 3. The Postgres Extensions: pgvecto.rs & Lantern\nIf you already use PostgreSQL and don\'t want to maintain a separate infrastructure just for vectors, Rust powers two major extensions:\n*   **[pgvecto.rs](embed:pgvecto-rs-Lg9)** (Apache 2.0): Focuses on scalable, low-latency hybrid search. It replaces standard Postgres indexing with highly optimized Rust bindings for ANN (Approximate Nearest Neighbor) searches.\n*   **[Lantern](embed:lantern-A4O)** (AGPL 3.0): Another Postgres extension heavily leaning into external index generation and edge use-cases.\n*   **Use Case:** Perfect if your embeddings are tightly coupled with existing relational data and you prefer scaling a single Postgres cluster.\n\n#### 4. The Specialized Engines: SeekStorm & Cozo\n*   **[SeekStorm](embed:seekstorm-p2o)**: Focuses heavily on combining traditional full-text (BM25/lexical) search with dense vector retrieval in a single query. It can run as an embedded in-process library or a standalone multi-tenant server.\n*   **[CozoDB](embed:cozo-nPL)**: A unique transactional database that uses **Datalog** (a declarative logic language) for querying. It brings together relational schemas, graph traversal, and vector search under one engine.\n\n### \u{1F4A1} Recommendation\n\n*   If you want the **safest, most robust, and widely supported** self-hosted engine: Choose **Qdrant**.\n*   If you are building an **advanced LLM/RAG pipeline** that relies on interconnected concepts: Choose **HelixDB**.\n*   If you want to keep everything inside **Postgres** but with better performance than standard `pgvector` (which is written in C): Choose **pgvecto.rs**.'
       }
     },
+    screenshot_to_html_pricing_card: {
+      title: {
+        text: "Screenshot to HTML pricing card"
+      },
+      summary: {
+        text: "Watch OpenMates turn a pricing-card screenshot into a standalone inline HTML code embed with visual QA metadata."
+      },
+      message_1: {
+        text: "Please turn this screenshot into a standalone inline HTML code embed. Match the screenshot closely and keep it self-contained.\n[!](embed:simple-card-input-png-fcfe81c4-5baa-4580-afa07b)"
+      },
+      message_2: {
+        text: '```json\n{"type": "app_skill_use", "embed_id": "395f8816-7e0b-4753-8058-f1f5c49d0482", "app_id": "code", "skill_id": "image_to_html"}\n```\n\n\u6211\u73B0\u5728\u6B63\u6839\u636E\u4F60\u7684\u622A\u56FE\u751F\u6210\u72EC\u7ACB\u7684 HTML \u4EE3\u7801 embed\u3002'
+      },
+      follow_up_1: {
+        text: "Run this HTML code to see how it works"
+      },
+      follow_up_2: {
+        text: "Explain how to customize this card's colors"
+      },
+      follow_up_3: {
+        text: "Find similar HTML card designs for inspiration"
+      },
+      follow_up_4: {
+        text: "Show me how to add interactive elements"
+      },
+      follow_up_5: {
+        text: "Generate a matching mobile version of this card"
+      },
+      follow_up_6: {
+        text: "Teach me about responsive design principles"
+      }
+    },
     sqlite_strict_tables_summary: {
       title: {
         text: "SQLite STRICT Tables for App Developers"

package/dist/cli.js

@@ -2,7 +2,7 @@
 import {
   getExtForLang,
   serializeToYaml
-} from "./chunk-LAAY2YYX.js";
+} from "./chunk-F2G7QKBM.js";
 import "./chunk-MLWBMJEN.js";
 export {
   getExtForLang,

package/dist/index.js

@@ -18,7 +18,7 @@ import {
   reconcileAuthoritativeChats,
   renderSupportInfo,
   serializeToYaml
-} from "./chunk-LAAY2YYX.js";
+} from "./chunk-F2G7QKBM.js";
 import "./chunk-MLWBMJEN.js";
 export {
   APP_SKILL_METADATA,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "openmates",
-  "version": "0.15.0-alpha.44",
+  "version": "0.15.0-alpha.45",
   "description": "OpenMates CLI and SDK",
   "type": "module",
   "main": "dist/index.js",