Updating google docs docs! (#558)

* Updating google docs docs!

* fixing broken link

---------

Co-authored-by: Francisco Liberal <[email protected]>

Author: jottakka

app/en/home/use-tools/tools-overview/page.mdx

@@ -13,7 +13,7 @@ To solve this, many AI models support tool calling (sometimes referred to as 'fu
 
 Say a colleague shares a document with you on Google Drive, and you'd like an LLM to help you analyze it.
 
-You could go to your Drive/Docs, open the document, copy its contents, and paste it into your chat. But what if the LLM could do this for you? The Arcade Google Docs MCP Server provides a [`SearchAndRetrieveDocuments`](/mcp-servers/productivity/google-docs#googledrivesearchandretrievedocuments) tool. By calling it, the LLM can find and read the document without you having to do anything.
+You could go to your Drive/Docs, open the document, copy its contents, and paste it into your chat. But what if the LLM could do this for you? The Arcade Google Docs MCP Server provides a [`SearchAndRetrieveDocuments`](/mcp-servers/productivity/google-docs#googledocssearchandretrievedocuments) tool. By calling it, the LLM can find and read the document without you having to do anything.
 
 After analyzing the document, you decide that a meeting is needed with your colleague. You can ask the LLM to schedule a meeting and it will use the [Google Calendar MCP Server](/mcp-servers/productivity/google-calendar) to do it without you needing to leave the chat.
 

app/en/mcp-servers/productivity/google-docs/page.mdx

@@ -30,6 +30,8 @@ import { Callout } from "nextra/components";
 The Arcade Google Docs MCP Server provides a pre-built set of tools for interacting with Google Docs. These tools make it easy to build agents and AI apps that can:
 
 - Create, update, list, and delete documents
+- Access document metadata including hierarchical tab structures
+- Retrieve document content with full tab support in multiple formats (Markdown, HTML, DocMD)
 
 ## Available Tools
 
@@ -47,6 +49,10 @@ These tools are currently available in the Arcade Google Docs MCP Sever.
       "GoogleDocs.GetDocumentAsDocMD",
       "Retrieve a Google Docs document by ID in DocMD format with metadata tags.",
     ],
+    [
+      "GoogleDocs.GetDocumentMetadata",
+      "Get metadata for a Google Docs document including hierarchical tab structure and approximate counts.",
+    ],
     [
       "GoogleDocs.EditDocument",
       "Edit a Google Docs document using natural language edit requests.",
@@ -98,10 +104,10 @@ These tools are currently available in the Arcade Google Docs MCP Sever.
       label: "Call the Tool with User Authorization",
       content: {
         Python: [
-          "/examples/integrations/mcp-servers/google_docs/who_am_i_example_call_tool.py",
+          "/examples/integrations/mcp-servers/google/docs/who_am_i_example_call_tool.py",
         ],
         JavaScript: [
-          "/examples/integrations/mcp-servers/google_docs/who_am_i_example_call_tool.js",
+          "/examples/integrations/mcp-servers/google/docs/who_am_i_example_call_tool.js",
         ],
       },
     },
@@ -152,21 +158,73 @@ Get the latest version of the specified Google Docs document.
       label: "Call the Tool with User Authorization",
       content: {
         Python: [
-          "/examples/integrations/mcp-servers/google/docs/get_document_as_docmd_example_call_tool.py",
+          "/examples/integrations/mcp-servers/google/docs/get_document_as_doc_md_example_call_tool.py",
         ],
         JavaScript: [
-          "/examples/integrations/mcp-servers/google/docs/get_document_as_docmd_example_call_tool.js",
+          "/examples/integrations/mcp-servers/google/docs/get_document_as_doc_md_example_call_tool.js",
         ],
       },
     },
   ]}
 />
 
-Get the latest version of the specified Google Docs document in DocMD format. The DocMD output includes tags that can be used to annotate the document with location information, block types, block IDs, and other metadata.
+Get the latest version of the specified Google Docs document in DocMD format. The DocMD output includes tags that can be used to annotate the document with location information, block types, block IDs, and other metadata. If the document has tabs, all tabs are included in sequential order unless a specific `tab_id` is provided.
 
 **Parameters**
 
 - **`document_id`** _(string, required)_ The ID of the document to retrieve.
+- **`tab_id`** _(string, optional)_ The ID of a specific tab to retrieve. If provided, returns only content from that tab. If omitted, returns all tabs in sequential depth-first order.
+
+---
+
+## GoogleDocs.GetDocumentMetadata
+
+<br />
+<TabbedCodeBlock
+  tabs={[
+    {
+      label: "Call the Tool with User Authorization",
+      content: {
+        Python: [
+          "/examples/integrations/mcp-servers/google/docs/get_document_metadata_example_call_tool.py",
+        ],
+        JavaScript: [
+          "/examples/integrations/mcp-servers/google/docs/get_document_metadata_example_call_tool.js",
+        ],
+      },
+    },
+  ]}
+/>
+
+Get metadata for a Google Docs document including hierarchical tab structure. Returns document title, ID, URL, approximate total character and word counts, and nested tab information with approximate character and word counts for each tab.
+
+**Parameters**
+
+- **`document_id`** _(string, required)_ The ID of the document to get metadata for.
+
+**Response Fields**
+
+The response includes the following fields:
+
+- **`documentId`** _(string)_ The unique identifier of the document.
+- **`title`** _(string)_ The title of the document.
+- **`documentUrl`** _(string)_ The URL to open and edit the document in Google Docs.
+- **`approximateTotalCharacterCount`** _(int)_ Approximate total number of characters across all tabs (or main body if no tabs).
+- **`approximateTotalWordCount`** _(int)_ Approximate total number of words across all tabs (or main body if no tabs).
+- **`tabsCount`** _(int)_ The total number of tabs in the document.
+- **`tabs`** _(list[dict])_ List of tabs with hierarchical structure. Each tab includes:
+  - **`tabId`** _(string)_ The unique identifier of the tab.
+  - **`title`** _(string)_ The title/name of the tab.
+  - **`index`** _(int)_ The position of the tab among its siblings (0-indexed).
+  - **`nestingLevel`** _(int)_ The nesting depth (0 for top-level, 1 for child, 2 for grandchild).
+  - **`approximateCharacterCount`** _(int)_ Approximate number of characters in this tab's content (excluding child tabs).
+  - **`approximateWordCount`** _(int)_ Approximate number of words in this tab's content (excluding child tabs).
+  - **`parentTabId`** _(string, optional)_ The ID of the parent tab (if this is a nested tab).
+  - **`childTabs`** _(list[dict], optional)_ List of nested child tabs within this tab.
+
+<Callout type="info">
+  **Note on Approximate Counts**: The character and word counts are approximate and may differ slightly from the exact counts shown in Google Docs. The counts are calculated by parsing the document structure and may not include all formatting or hidden content.
+</Callout>
 
 ---
 
@@ -316,7 +374,7 @@ Search Google documents in the user's Google Drive. Excludes documents that are
 - **`limit`** _(int, optional)_ The number of documents to list. Defaults to `50`.
 - **`pagination_token`** _(str, optional)_ The pagination token to continue a previous request
 
-## GoogleDrive.SearchAndRetrieveDocuments
+## GoogleDocs.SearchAndRetrieveDocuments
 
 <br />
 <TabbedCodeBlock
@@ -335,11 +393,11 @@ Search Google documents in the user's Google Drive. Excludes documents that are
   ]}
 />
 
-Searches for documents in the user's Google Drive and returns a list of documents (with text content) matching the search criteria. Excludes documents that are in the trash.
+Searches for documents in the user's Google Drive and returns a list of documents (with text content) matching the search criteria. Excludes documents that are in the trash. If the document has tabs, all tab content is included in the output.
 
 **Parameters**
 
-- **`document_format`** _(enum ([DocumentFormat](#documentformat)), optional)_ The format of the document to be returned. Defaults to Markdown.
+- **`return_format`** _(enum ([DocumentFormat](#documentformat)), optional)_ The format of the document to be returned. Defaults to Markdown. Can be `MARKDOWN`, `HTML`, `DOCMD`, or `GOOGLE_API_JSON`.
 - **`document_contains`** _(list[str], optional)_ Keywords or phrases that must be in the document title or body. Provide a list of keywords or phrases if needed.
 - **`document_not_contains`** _(list[str], optional)_ Keywords or phrases that must not be in the document title or body. Provide a list of keywords or phrases if needed.
 - **`search_only_in_shared_drive_id`** _(str, optional)_ The ID of the shared drive to restrict the search to. If provided, the search will only return documents from this drive. Defaults to None, which searches across all drives.
@@ -407,6 +465,21 @@ Comment on a specific document by its ID.
 
 ---
 
+## Tab Support
+
+Google Docs supports hierarchical tabs within documents. The Google Docs tools provide comprehensive support for working with tabs:
+
+- **Tab Metadata**: `GetDocumentMetadata` returns hierarchical tab structures with approximate character and word counts for each tab
+- **Tab Content**: `GetDocumentAsDocMD` and `SearchAndRetrieveDocuments` include all tab content in their output
+- **Tab Filtering**: `GetDocumentAsDocMD` supports filtering to retrieve content from a specific tab using the `tab_id` parameter
+
+Tabs are represented with the following structure:
+- Each tab has a unique `tabId`, `title`, `index`, and `nestingLevel`
+- Tabs can be nested up to 3 levels deep (parent → child → grandchild)
+- Tab metadata includes approximate character and word counts for each tab's content
+
+---
+
 ## Auth
 
 The Arcade Docs MCP Sever uses the [Google auth provider](/home/auth-providers/google) to connect to users' Google accounts.
@@ -423,8 +496,9 @@ With a self-hosted installation of Arcade, you need to [configure the Google aut
 
 The format of the document to be returned.
 
-- **`MARKDOWN`**: Markdown format.
-- **`HTML`**: HTML format.
+- **`MARKDOWN`**: Markdown format. Includes all tabs in the output if present.
+- **`HTML`**: HTML format. Includes all tabs in the output if present.
+- **`DOCMD`**: DocMD format with metadata tags. Includes all tabs in the output if present.
 - **`GOOGLE_API_JSON`**: Original JSON format returned by the Google API.
 
 ### OrderBy

public/examples/integrations/mcp-servers/google/docs/get_document_metadata_example_call_tool.js

@@ -0,0 +1,32 @@
+import { Arcade } from "@arcadeai/arcadejs";
+
+const client = new Arcade(); // Automatically finds the `ARCADE_API_KEY` env variable
+
+const USER_ID = "{arcade_user_id}";
+const TOOL_NAME = "GoogleDocs.GetDocumentMetadata";
+
+// Start the authorization process
+const authResponse = await client.tools.authorize({
+  tool_name: TOOL_NAME,
+  user_id: USER_ID,
+});
+
+if (authResponse.status !== "completed") {
+  console.log(`Click this link to authorize: ${authResponse.url}`);
+}
+
+// Wait for the authorization to complete
+await client.auth.waitForCompletion(authResponse);
+
+const toolInput = {
+  document_id: "your_document_id_here" // Document Ids can be found with the SearchDocuments tool
+};
+
+const response = await client.tools.execute({
+  tool_name: TOOL_NAME,
+  input: toolInput,
+  user_id: USER_ID,
+});
+
+console.log(response);
+

public/examples/integrations/mcp-servers/google/docs/get_document_metadata_example_call_tool.py

@@ -0,0 +1,29 @@
+from arcadepy import Arcade
+
+client = Arcade()  # Automatically finds the `ARCADE_API_KEY` env variable
+
+USER_ID = "{arcade_user_id}"
+TOOL_NAME = "GoogleDocs.GetDocumentMetadata"
+
+auth_response = client.tools.authorize(
+    tool_name=TOOL_NAME,
+    user_id=USER_ID,
+)
+
+if auth_response.status != "completed":
+    print(f"Click this link to authorize: {auth_response.url}")
+
+# Wait for the authorization to complete
+client.auth.wait_for_completion(auth_response)
+
+tool_input = {
+    "document_id": "your_document_id_here"  # Document Ids can be found with the SearchDocuments tool
+}
+
+response = client.tools.execute(
+    tool_name=TOOL_NAME,
+    input=tool_input,
+    user_id=USER_ID,
+)
+print(response)
+

public/examples/integrations/mcp-servers/google/docs/who_am_i_example_call_tool.js

@@ -0,0 +1,27 @@
+import { Arcade } from "@arcadeai/arcadejs";
+
+const client = new Arcade(); // Automatically finds the `ARCADE_API_KEY` env variable
+
+const USER_ID = "{arcade_user_id}";
+const TOOL_NAME = "GoogleDocs.WhoAmI";
+
+// Start the authorization process
+const authResponse = await client.tools.authorize({tool_name: TOOL_NAME});
+
+if (authResponse.status !== "completed") {
+  console.log(`Click this link to authorize: ${authResponse.url}`);
+}
+
+// Wait for the authorization to complete
+await client.auth.waitForCompletion(authResponse);
+
+const toolInput = {};
+
+const response = await client.tools.execute({
+  tool_name: TOOL_NAME,
+  input: toolInput,
+  user_id: USER_ID,
+});
+
+console.log(JSON.stringify(response.output.value, null, 2));
+

public/examples/integrations/mcp-servers/google/docs/who_am_i_example_call_tool.py

@@ -0,0 +1,27 @@
+import json
+from arcadepy import Arcade
+
+client = Arcade()  # Automatically finds the `ARCADE_API_KEY` env variable
+
+USER_ID = "{arcade_user_id}"
+TOOL_NAME = "GoogleDocs.WhoAmI"
+
+auth_response = client.tools.authorize(tool_name=TOOL_NAME)
+
+if auth_response.status != "completed":
+    print(f"Click this link to authorize: {auth_response.url}")
+
+# Wait for the authorization to complete
+client.auth.wait_for_completion(auth_response)
+
+tool_input = {
+    
+}
+
+response = client.tools.execute(
+    tool_name=TOOL_NAME,
+    input=tool_input,
+    user_id=USER_ID,
+)
+print(json.dumps(response.output.value, indent=2))
+