Update mcp-integration.adoc

Author: ShashiSubramanya

modules/ROOT/pages/mcp-integration.adoc

@@ -40,13 +40,15 @@ The `getRelevantQuestions` tool to fetch relevant data questions for a given dat
 * `getAnswer` to execute the queries and fetch data +
 The `getAnswer` tool generates answers and insights for a given data context.
 * `createLiveboard` to create a Liveboard in ThoughtSpot +
-The `createLiveboard` tool calls the Liveboard creation workflow and creates a Liveboard with the answers generated from user's query.
+The `createLiveboard` tool calls the Liveboard creation workflow and creates a Liveboard with the answers generated from the user's query.
+
+////
 * `getDataSourceSuggestions` to get data source suggestions +
 Based on the type of data that users want to fetch, `getDataSourceSuggestions` gets a list of data source recommendations. Currently, `getDataSourceSuggestions` is not exposed as an MCP tool and is available as an MCP `resource`. To get data source suggestions, the user or MCP client must have at least view access to ThoughtSpot data sources.
-
+////
 
 MCP client/ LLM agent::
-The external system or application environment with AI Agent, Claude, OpenAI, or a custom chatbot that acts as a user interface and  orchestrates interaction with ThoughtSpot MCP Server
+The external system or application environment with AI Agent, Claude, OpenAI, or a custom chatbot that acts as a user interface and  orchestrates interaction with the ThoughtSpot MCP Server.
 This is the model or system that processes the user’s natural language input, determines which tool to call, and integrates the tool results into its final output.
 
 ////
@@ -57,11 +59,11 @@ Integration requires configuration, typically via a config file, to specify serv
 Authentication and security settings::
 
 * Access to ThoughtSpot instance: +
-For MCP Server connection, users require access to a ThoughtSpot instance. For tool invocation, the MCP server must accept authenticated requests and the LLM tool specification must carry those credentials or headers. +
-ThoughtSpot administrators can use the SSO framework with SAML or OAuth token-based authentication methods to authenticate and sign in the users. +
+For MCP Server connection, users require access to a ThoughtSpot instance. For tool invocation, the MCP server must accept authenticated requests, and the LLM tool specification must carry those credentials or headers. +
+ThoughtSpot administrators can use the SSO framework with SAML or OAuth token-based authentication methods to authenticate and sign in users. +
 * SAML redirect settings: +
 For SAML SSO users, the SAML redirect domain configuration is required to ensure that users are redirected to an allowed and trusted domain after they are authenticated. +
-* To get answers to their data queries, your application users require at least view access to ThoughtSpot data sources. To generate an Answer or to create Liveboard, users require data download privilege.
+* To get answers to their data queries, your application users require at least view access to ThoughtSpot data sources. To generate an Answer or to create a Liveboard, users require the data download privilege.
 * CSP and CORS settings: +
 To secure communication between the MCP client and the ThoughtSpot instance, administrators must add the MCP Server URL to CSP (Content Security Policy) and CORS (Cross-Origin Resource Sharing) allowlists in ThoughtSpot.
 * Client connection configuration: +
@@ -75,7 +77,7 @@ The MCP Server integration with an agentic framework or LLM clients enables the
 . User sends a query to get data from a specific ThoughtSpot data model context.
 . The LLM / AI agent receives the request and sends it to the MCP server endpoint with the user's query.
 . The MCP server responds with the available tools.
-. The LLM / AI Agent determines the appropriate MCP tool to call. Based on the user's query or prompt, the MCP tools are invoked. For example, to get information for a specific data context from ThoughtSpot, break down the user's query into relevant questions or create an artifact in ThoughtSpot programmatically.
+. The LLM / AI Agent determines the appropriate MCP tool to call. Based on the user's query or prompt, the MCP tools are invoked. For example, to get information for a specific data context from ThoughtSpot, break down the user's query into relevant questions or programmatically create an artifact in ThoughtSpot.
 . The MCP server processes the request and returns the result.
 . The agent receives the response, constructs the output, and presents it to the user.
 . User receives the response. The user can refine the analysis with follow-up queries for further exploration or ask a new question. +
@@ -107,7 +109,7 @@ To enable secure communication between the MCP Server and your ThoughtSpot insta
 ////
 === Configure security settings on ThoughtSpot
 
-To allow the secure communication between the MCP Server and your ThoughtSpot instance, configure the following settings:
+To allow secure communication between the MCP Server and your ThoughtSpot instance, configure the following settings:
 
 . On your ThoughtSpot instance, navigate to *Develop* > *Customizations* > *Security Settings*.
 . Add the MCP Server domain to CSP and CORS allowlists.
@@ -116,7 +118,7 @@ To allow the secure communication between the MCP Server and your ThoughtSpot in
 
 === Connect your client to the MCP Server
 
-If using a client that supports remote MCPs natively such as Claude AI, use the following MCP server URL:
+If using a client that supports remote MCPs natively, such as Claude AI, use the following MCP server URL:
 ----
 https://agent.thoughtspot.app/mcp
 ----
@@ -126,11 +128,11 @@ For OpenAI ChatGPT Deep Research, use the following URL:
 https://agent.thoughtspot.app/openai/mcp
 ----
 
-For MCP clients such as Claude Desktop, Windsurf, Cursor, that do not support remote MCP Server, you must xref:mcp-integration.adoc#_connecting_other_mcp_clients_claude_desktop[add the MCP server configuration to your MCP client settings].
+For MCP clients such as Claude Desktop, Windsurf, and Cursor that do not support a remote MCP Server, you must xref:mcp-integration.adoc#_connecting_other_mcp_clients_claude_desktop[add the MCP server configuration to your MCP client settings].
 
 === Call MCP tools via LLM APIs
 
-ThoughtSpot remote MCP Server acts as a wrapper over the ThoughtSpot APIs, making them available as tools for agent frameworks or LLMs such as Claude or OpenAI. It exposes specific tools to get relevant questions, answer, datasource suggestions, or create a Liveboard, which can be invoked by the LLMs in response to a user's query or prompt.
+ThoughtSpot remote MCP Server acts as a wrapper over the ThoughtSpot APIs, making them available as tools for agent frameworks or LLMs such as Claude or OpenAI. It exposes specific tools that can be invoked by the LLMs in response to a user's query or prompt.
 
 To enable tool calling:
 
@@ -148,7 +150,7 @@ For information about calling MCP tools using LLM APIs and methods, see these se
 * xref:mcp-integration.adoc#_gemini_api[Gemini API and function calling]
 
 ==== Claude MCP connector
-Claude’s MCP connector feature enables you to connect to remote MCP Servers directly from the Messages API.
+The Claude’s MCP connector allows you to connect to remote MCP Servers directly from the Messages API.
 
 To connect to the ThoughtSpot remote MCP Server, specify the following properties in the API request:
 
@@ -161,7 +163,7 @@ __String__. The URL of the remote MCP Server endpoint. Must start with `https://
 ** `name` +
 __String__. A unique identifier/label for the MCP Server. It will be used in the MCP tool call blocks to identify the server and to disambiguate tools to the LLM.
 ** `authorization_token` +
-__String__. OAuth authorization token (`TS_AUTH_TOKEN`) along with the ThoughtSpot application instance URL. In the following  example, authorization token is prefixed, and the ThoughtSpot host URL is added with the `@` symbol.
+__String__. OAuth authorization token (`TS_AUTH_TOKEN`) along with the ThoughtSpot application instance URL. In the following example, the authorization token is added as a prefix, and the ThoughtSpot host URL is added with the `@` symbol.
 
 * `messages` +
 In the `messages` array, specify a natural language question in `content` and the user role in `role`.
@@ -324,7 +326,7 @@ const response = await ai.models.generateContent({
 });
 console.log(response.text)
 
-// Close the connection
+// Close the connection,
 await client.close();
 ----
 
@@ -336,7 +338,7 @@ For additional information, refer to the following resources:
 
 === For clients that do not support the remote MCP server
 
-For clients such as Claude Desktop, Windsurf, Cursor, which do not support remote MCP servers, add the following configuration to your MCP client settings:
+For clients such as Claude Desktop, Windsurf, and Cursor, which do not support remote MCP servers, add the following configuration to your MCP client settings:
 
 [source,JSON]
 ----