fix(firecrawl,daytona): export the search envelope consts, correct the toolbox TSDoc

The generated firecrawl doc hoisted the web item's fields onto data itself,
because the docs generator matches 'properties: CONST' at any depth and hit
the nested items. Splitting the envelope into an exported *_OUTPUT_PROPERTIES
const matches the pattern the rest of the file already uses.

Daytona's daytonaToolboxUrl now records why every call site omits baseUrl: the
request carries the bearer token, so the host must never come from a caller,
and a real fix needs a server-side lookup.

Author: waleedlatif1

apps/docs/content/docs/en/integrations/firecrawl.mdx

@@ -176,9 +176,38 @@ Search for information on the web using Firecrawl
 | Parameter | Type | Description |
 | --------- | ---- | ----------- |
 | `data` | object | Search results, grouped by the sources requested |
-|   ↳ `title` | string | Title from the search result |
-|   ↳ `description` | string | Description from the search result |
-|   ↳ `url` | string | URL of the search result |
+|   ↳ `web` | array | Web results \(returned by default, or when "web" is in sources\) |
+|     ↳ `markdown` | string | Page content in markdown \(when scrapeOptions.formats includes "markdown"\) |
+|     ↳ `html` | string | Processed HTML content \(when scrapeOptions.formats includes "html"\) |
+|     ↳ `rawHtml` | string | Unprocessed raw HTML \(when scrapeOptions.formats includes "rawHtml"\) |
+|     ↳ `links` | array | Links found on the page \(when scrapeOptions.formats includes "links"\) |
+|     ↳ `screenshot` | string | Screenshot URL \(expires after 24 hours, when scrapeOptions.formats includes "screenshot"\) |
+|     ↳ `audio` | string | Signed URL to the extracted MP3 \(when scrapeOptions.formats includes "audio"\); expires after 1 hour |
+|     ↳ `video` | string | Signed URL to the extracted video \(when scrapeOptions.formats includes "video"\); expires after 1 hour |
+|     ↳ `title` | string | Title from the search result |
+|     ↳ `description` | string | Description from the search result |
+|     ↳ `url` | string | URL of the search result |
+|   ↳ `news` | array | News results \(only when "news" is in sources\) |
+|     ↳ `markdown` | string | Page content in markdown \(when scrapeOptions.formats includes "markdown"\) |
+|     ↳ `html` | string | Processed HTML content \(when scrapeOptions.formats includes "html"\) |
+|     ↳ `rawHtml` | string | Unprocessed raw HTML \(when scrapeOptions.formats includes "rawHtml"\) |
+|     ↳ `links` | array | Links found on the page \(when scrapeOptions.formats includes "links"\) |
+|     ↳ `screenshot` | string | Screenshot URL \(expires after 24 hours, when scrapeOptions.formats includes "screenshot"\) |
+|     ↳ `audio` | string | Signed URL to the extracted MP3 \(when scrapeOptions.formats includes "audio"\); expires after 1 hour |
+|     ↳ `video` | string | Signed URL to the extracted video \(when scrapeOptions.formats includes "video"\); expires after 1 hour |
+|     ↳ `title` | string | Title of the article |
+|     ↳ `snippet` | string | Snippet from the article |
+|     ↳ `url` | string | URL of the article |
+|     ↳ `date` | string | Date of the article |
+|     ↳ `imageUrl` | string | URL of the article's image |
+|     ↳ `position` | number | Rank of the article within the news results \(1-based\) |
+|   ↳ `images` | array | Image results \(only when "images" is in sources\) |
+|     ↳ `title` | string | Title from the search result |
+|     ↳ `imageUrl` | string | URL of the image itself |
+|     ↳ `imageWidth` | number | Width of the image in pixels |
+|     ↳ `imageHeight` | number | Height of the image in pixels |
+|     ↳ `url` | string | URL of the page containing the image |
+|     ↳ `position` | number | Rank of the result within the image results \(1-based\) |
 | `warning` | string | Warning message if any issues occurred during the search |
 | `id` | string | ID of the search job |
 | `creditsUsed` | number | Number of credits the search consumed |

apps/sim/tools/firecrawl/types.ts

@@ -149,7 +149,7 @@ export const CRAWLED_PAGE_OUTPUT_PROPERTIES = {
  *
  * Based on the `data.web[]` / `data.news[]` items of POST /v2/search.
  */
-const SEARCH_SCRAPED_OUTPUT_PROPERTIES = {
+export const SEARCH_SCRAPED_OUTPUT_PROPERTIES = {
   markdown: {
     type: 'string',
     description: 'Page content in markdown (when scrapeOptions.formats includes "markdown")',
@@ -252,29 +252,32 @@ export const SEARCH_IMAGE_RESULT_OUTPUT_PROPERTIES = {
  * an image `imageUrl` have no equivalent on a web result and would be lost if
  * the sources were merged.
  */
+export const SEARCH_DATA_OUTPUT_PROPERTIES = {
+  web: {
+    type: 'array',
+    description: 'Web results (returned by default, or when "web" is in sources)',
+    optional: true,
+    items: { type: 'object', properties: SEARCH_WEB_RESULT_OUTPUT_PROPERTIES },
+  },
+  news: {
+    type: 'array',
+    description: 'News results (only when "news" is in sources)',
+    optional: true,
+    items: { type: 'object', properties: SEARCH_NEWS_RESULT_OUTPUT_PROPERTIES },
+  },
+  images: {
+    type: 'array',
+    description: 'Image results (only when "images" is in sources)',
+    optional: true,
+    items: { type: 'object', properties: SEARCH_IMAGE_RESULT_OUTPUT_PROPERTIES },
+  },
+} as const satisfies Record<string, OutputProperty>
+
+/** The declared `data` output of POST /v2/search. */
 export const SEARCH_DATA_OUTPUT: OutputProperty = {
   type: 'object',
   description: 'Search results, grouped by the sources requested',
-  properties: {
-    web: {
-      type: 'array',
-      description: 'Web results (returned by default, or when "web" is in sources)',
-      optional: true,
-      items: { type: 'object', properties: SEARCH_WEB_RESULT_OUTPUT_PROPERTIES },
-    },
-    news: {
-      type: 'array',
-      description: 'News results (only when "news" is in sources)',
-      optional: true,
-      items: { type: 'object', properties: SEARCH_NEWS_RESULT_OUTPUT_PROPERTIES },
-    },
-    images: {
-      type: 'array',
-      description: 'Image results (only when "images" is in sources)',
-      optional: true,
-      items: { type: 'object', properties: SEARCH_IMAGE_RESULT_OUTPUT_PROPERTIES },
-    },
-  },
+  properties: SEARCH_DATA_OUTPUT_PROPERTIES,
 }
 
 /** Declared shape of one `MapDocument` entry in a map result. */