Adding Grok provider

package.json

@@ -73,6 +73,7 @@
         "@langchain/mistralai": "^0.2.1",
         "@langchain/ollama": "^0.2.3",
         "@langchain/openai": "^0.5.16",
+        "@langchain/xai": "^0.0.3",
         "@lumino/coreutils": "^2.1.2",
         "@lumino/polling": "^2.1.2",
         "@lumino/signaling": "^2.1.2",

scripts/settings-checker.js

@@ -69,6 +69,10 @@ const providers = {
     type: 'ChatOpenAIFields',
     excludedProps: ['configuration']
   },
+  Grok: {
+    path: 'node_modules/@langchain/xai/dist/chat_models.d.ts',
+    type: 'ChatXAIInput'
+  },
   WebLLM: {
     path: 'node_modules/@langchain/community/chat_models/webllm.d.ts',
     type: 'WebLLMInputs',

src/default-providers/Grok/completer.ts

@@ -0,0 +1,60 @@
+import {
+  CompletionHandler,
+  IInlineCompletionContext
+} from '@jupyterlab/completer';
+import { ChatXAI } from '@langchain/xai';
+import { AIMessage, SystemMessage } from '@langchain/core/messages';
+
+import { BaseCompleter } from '../../base-completer';
+
+export class GrokCompleter extends BaseCompleter {
+  constructor(options: BaseCompleter.IOptions) {
+    super(options);
+
+    this._completer = new ChatXAI({
+      model: 'grok-4',
+      ...options.settings
+    });
+  }
+
+  async fetch(
+    request: CompletionHandler.IRequest,
+    context: IInlineCompletionContext
+  ) {
+    const { text, offset: cursorOffset } = request;
+    const prompt = text.slice(0, cursorOffset).trim();
+
+    const messages = [
+      new SystemMessage(this.systemPrompt),
+      new AIMessage(prompt)
+    ];
+
+    try {
+      const response = await this._completer.invoke(messages);
+      const items = [];
+      if (typeof response.content === 'string') {
+        items.push({
+          insertText: response.content
+        });
+      } else {
+        response.content.forEach(content => {
+          if (content.type !== 'text') {
+            return;
+          }
+          items.push({
+            insertText: content.text,
+            filterText: prompt.substring(prompt.length)
+          });
+        });
+      }
+      return {
+        items
+      };
+    } catch (error) {
+      console.error('Error fetching Grok completions', error);
+      return { items: [] };
+    }
+  }
+
+  protected _completer: ChatXAI;
+}

src/default-providers/Grok/instructions.ts

@@ -0,0 +1,11 @@
+export default `
+<i class="fas fa-exclamation-triangle"></i> This extension is experimental and not affiliated with xAI.
+
+1. Visit <https://console.x.ai/> and sign in to access your API key.
+2. In JupyterLab, go to **Settings → AI Providers → Grok**.
+3. Enter your API key in the \`apiKey\` field (required).
+4. Choose a model like \`grok-4\` (default).
+5. Open the chat or use inline completions to interact with Grok.
+
+**Note:** *This extension uses the xAI API to provide AI capabilities. Ensure you have an API key (with credits) to use this extension.*
+`;

src/default-providers/Grok/settings-schema.json

@@ -0,0 +1,25 @@
+{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "type": "object",
+  "properties": {
+    "apiKey": {
+      "type": "string",
+      "description": "Your xAI Grok API Key"
+    },
+    "model": {
+      "type": "string",
+      "description": "Model name to use (e.g. grok-4)",
+      "default": "grok-4"
+    },
+    "temperature": {
+      "type": "number",
+      "description": "Sampling temperature",
+      "default": 0.7
+    },
+    "maxTokens": {
+      "type": "number",
+      "description": "Maximum number of tokens to generate"
+    }
+  },
+  "additionalProperties": false
+}

src/default-providers/index.ts

@@ -11,6 +11,7 @@ import { ChatGoogleGenerativeAI } from '@langchain/google-genai';
 import { ChatMistralAI } from '@langchain/mistralai';
 import { ChatOllama } from '@langchain/ollama';
 import { ChatOpenAI } from '@langchain/openai';
+import { ChatXAI } from '@langchain/xai';
 
 // Import completers
 import { AnthropicCompleter } from './Anthropic/completer';
@@ -20,6 +21,7 @@ import { CodestralCompleter } from './MistralAI/completer';
 import { OllamaCompleter } from './Ollama/completer';
 import { OpenAICompleter } from './OpenAI/completer';
 import { WebLLMCompleter } from './WebLLM/completer';
+import { GrokCompleter } from './Grok/completer';
 
 // Import Settings
 import AnthropicSettings from './Anthropic/settings-schema.json';
@@ -29,6 +31,7 @@ import MistralAISettings from './MistralAI/settings-schema.json';
 import OllamaAISettings from './Ollama/settings-schema.json';
 import OpenAISettings from './OpenAI/settings-schema.json';
 import WebLLMSettings from './WebLLM/settings-schema.json';
+import GrokSettings from './Grok/settings-schema.json';
 
 // Import instructions
 import ChromeAIInstructions, {
@@ -40,6 +43,7 @@ import OllamaInstructions from './Ollama/instructions';
 import WebLLMInstructions, {
   compatibilityCheck as webLLMCompatibilityCheck
 } from './WebLLM/instructions';
+import GrokInstructions from './Grok/instructions';
 
 import { prebuiltAppConfig } from '@mlc-ai/web-llm';
 
@@ -90,6 +94,13 @@ const AIProviders: IAIProvider[] = [
     chat: ChatOpenAI,
     completer: OpenAICompleter,
     settingsSchema: OpenAISettings
+  },
+  {
+    name: 'Grok',
+    chat: ChatXAI,
+    completer: GrokCompleter,
+    settingsSchema: GrokSettings,
+    instructions: GrokInstructions
   }
 ];