feat: add html scrape support

server/src/api/record.ts

@@ -18,7 +18,7 @@ import { WorkflowFile } from "maxun-core";
 import { googleSheetUpdateTasks, processGoogleSheetUpdates } from "../workflow-management/integrations/gsheet";
 import { airtableUpdateTasks, processAirtableUpdates } from "../workflow-management/integrations/airtable";
 import { sendWebhook } from "../routes/webhook";
-import { convertPageToMarkdown } from '../markdownify/scrape';
+import { convertPageToHTML, convertPageToMarkdown } from '../markdownify/scrape';
 
 chromium.use(stealthPlugin());
 
@@ -346,7 +346,8 @@ function formatRunResponse(run: any) {
         data: {
             textData: {},
             listData: {},
-            markdown: ''
+            markdown: '',
+            html: ''
         },
         screenshots: [] as any[],
     };
@@ -365,6 +366,10 @@ function formatRunResponse(run: any) {
         formattedRun.data.markdown = output.markdown[0]?.content || '';
     }
 
+    if (output.html && Array.isArray(output.html)) {
+        formattedRun.data.html = output.html[0]?.content || '';
+    }
+
     if (run.binaryOutput) {
         Object.keys(run.binaryOutput).forEach(key => {
             if (run.binaryOutput[key]) {
@@ -575,9 +580,9 @@ async function triggerIntegrationUpdates(runId: string, robotMetaId: string): Pr
   }
 }
 
-async function readyForRunHandler(browserId: string, id: string, userId: string){
+async function readyForRunHandler(browserId: string, id: string, userId: string, requestedFormats?: string[]){
     try {
-        const result = await executeRun(id, userId);
+        const result = await executeRun(id, userId, requestedFormats);
 
         if (result && result.success) {
             logger.log('info', `Interpretation of ${id} succeeded`);
@@ -614,7 +619,7 @@ function AddGeneratedFlags(workflow: WorkflowFile) {
     return copy;
 };
 
-async function executeRun(id: string, userId: string) {
+async function executeRun(id: string, userId: string, requestedFormats?: string[]) {
     let browser: any = null;
     
     try {
@@ -657,12 +662,19 @@ async function executeRun(id: string, userId: string) {
             };
         }
 
-        if (recording.recording_meta.type === 'markdown') {
+        if (recording.recording_meta.type === 'scrape') {
-            logger.log('info', `Executing markdown robot for API run ${id}`);
+            logger.log('info', `Executing scrape robot for API run ${id}`);
+
+            let formats = recording.recording_meta.formats || ['markdown'];
+
+            // Override if API request defines formats
+            if (requestedFormats && Array.isArray(requestedFormats) && requestedFormats.length > 0) {
+                formats = requestedFormats.filter((f): f is 'markdown' | 'html' => ['markdown', 'html'].includes(f));
+            }
 
             await run.update({
                 status: 'running',
-                log: 'Converting page to markdown'
+                log: `Converting page to: ${formats.join(', ')}`
             });
 
             try {
@@ -672,20 +684,33 @@ async function executeRun(id: string, userId: string) {
                     throw new Error('No URL specified for markdown robot');
                 }
 
-                const markdown = await convertPageToMarkdown(url);
+                let markdown = '';
+                let html = '';
+                const serializableOutput: any = {};
+
+                // Markdown conversion
+                if (formats.includes('markdown')) {
+                    markdown = await convertPageToMarkdown(url);
+                    serializableOutput.markdown = [{ content: markdown }];
+                }
+
+                // HTML conversion
+                if (formats.includes('html')) {
+                    html = await convertPageToHTML(url);
+                    serializableOutput.html = [{ content: html }];
+                }
 
                 await run.update({
                     status: 'success',
                     finishedAt: new Date().toLocaleString(),
-                    log: 'Markdown conversion completed successfully',
+                    log: `${formats.join(', ')} conversion completed successfully`,
-                    serializableOutput: {
+                    serializableOutput,
-                        markdown: [{ content: markdown }]
-                    },
                     binaryOutput: {},
                 });
 
                 logger.log('info', `Markdown robot execution completed for API run ${id}`);
 
+                // Push success socket event
                 try {
                     const completionData = {
                         runId: plainRun.runId,
@@ -695,30 +720,45 @@ async function executeRun(id: string, userId: string) {
                         finishedAt: new Date().toLocaleString()
                     };
 
-                    serverIo.of('/queued-run').to(`user-${userId}`).emit('run-completed', completionData);
+                    serverIo
+                        .of('/queued-run')
+                        .to(`user-${userId}`)
+                        .emit('run-completed', completionData);
                 } catch (socketError: any) {
-                    logger.log('warn', `Failed to send run-completed notification for markdown robot run ${id}: ${socketError.message}`);
+                    logger.log(
+                        'warn',
+                        `Failed to send run-completed notification for markdown robot run ${id}: ${socketError.message}`
+                    );
                 }
 
-                const webhookPayload = {
+                // Build webhook payload
+                const webhookPayload: any = {
                     robot_id: plainRun.robotMetaId,
                     run_id: plainRun.runId,
                     robot_name: recording.recording_meta.name,
                     status: 'success',
                     started_at: plainRun.startedAt,
                     finished_at: new Date().toLocaleString(),
-                    markdown: markdown,
                     metadata: {
                         browser_id: plainRun.browserId,
                         user_id: userId,
-                    }
+                    },
                 };
 
+                if (formats.includes('markdown')) webhookPayload.markdown = markdown;
+                if (formats.includes('html')) webhookPayload.html = html;
+
                 try {
                     await sendWebhook(plainRun.robotMetaId, 'run_completed', webhookPayload);
-                    logger.log('info', `Webhooks sent successfully for markdown robot API run ${plainRun.runId}`);
+                    logger.log(
+                        'info',
+                        `Webhooks sent successfully for markdown robot API run ${plainRun.runId}`
+                    );
                 } catch (webhookError: any) {
-                    logger.log('warn', `Failed to send webhooks for markdown robot run ${plainRun.runId}: ${webhookError.message}`);
+                    logger.log(
+                        'warn',
+                        `Failed to send webhooks for markdown robot run ${plainRun.runId}: ${webhookError.message}`
+                    );
                 }
 
                 await destroyRemoteBrowser(plainRun.browserId, userId);
@@ -728,14 +768,18 @@ async function executeRun(id: string, userId: string) {
                     interpretationInfo: run.toJSON()
                 };
             } catch (error: any) {
-                logger.log('error', `Markdown conversion failed for API run ${id}: ${error.message}`);
+                logger.log(
+                    'error',
+                    `${formats.join(', ')} conversion failed for API run ${id}: ${error.message}`
+                );
 
                 await run.update({
                     status: 'failed',
                     finishedAt: new Date().toLocaleString(),
-                    log: `Markdown conversion failed: ${error.message}`,
+                    log: `${formats.join(', ')} conversion failed: ${error.message}`,
                 });
 
+                // Send failure socket event
                 try {
                     const failureData = {
                         runId: plainRun.runId,
@@ -745,9 +789,15 @@ async function executeRun(id: string, userId: string) {
                         finishedAt: new Date().toLocaleString()
                     };
 
-                    serverIo.of('/queued-run').to(`user-${userId}`).emit('run-completed', failureData);
+                    serverIo
+                        .of('/queued-run')
+                        .to(`user-${userId}`)
+                        .emit('run-completed', failureData);
                 } catch (socketError: any) {
-                    logger.log('warn', `Failed to send run-failed notification for markdown robot run ${id}: ${socketError.message}`);
+                    logger.log(
+                        'warn',
+                        `Failed to send run-failed notification for markdown robot run ${id}: ${socketError.message}`
+                    );
                 }
 
                 await destroyRemoteBrowser(plainRun.browserId, userId);
@@ -953,7 +1003,7 @@ async function executeRun(id: string, userId: string) {
     }
 }
 
-export async function handleRunRecording(id: string, userId: string) {
+export async function handleRunRecording(id: string, userId: string, requestedFormats?: string[]) {
     try {
         const result = await createWorkflowAndStoreMetadata(id, userId);
         const { browserId, runId: newRunId } = result;
@@ -967,7 +1017,7 @@ export async function handleRunRecording(id: string, userId: string) {
             rejectUnauthorized: false
         });
 
-        socket.on('ready-for-run', () => readyForRunHandler(browserId, newRunId, userId));
+        socket.on('ready-for-run', () => readyForRunHandler(browserId, newRunId, userId, requestedFormats));
 
         logger.log('info', `Running Robot: ${id}`);
 
@@ -1018,6 +1068,21 @@ async function waitForRunCompletion(runId: string, interval: number = 2000) {
  *           type: string
  *         required: true
  *         description: The ID of the robot to run.
+ *     requestBody:
+ *       required: false
+ *       content:
+ *         application/json:
+ *           schema:
+ *             type: object
+ *             properties:
+ *               formats:
+ *                 type: array
+ *                 items:
+ *                   type: string
+ *                   enum: [markdown, html]
+ *                 description: Optional override formats for this run.
+ *           example:
+ *             formats: ["html"]
  *     responses:
  *       200:
  *         description: Robot run started successfully.
@@ -1076,7 +1141,10 @@ router.post("/robots/:id/runs", requireAPIKey, async (req: AuthenticatedRequest,
         if (!req.user) {
             return res.status(401).json({ ok: false, error: 'Unauthorized' });
         }
-        const runId = await handleRunRecording(req.params.id, req.user.id);
+
+        const requestedFormats = req.body.formats;
+
+        const runId = await handleRunRecording(req.params.id, req.user.id, requestedFormats);
 
         if (!runId) {
             throw new Error('Run ID is undefined');

server/src/markdownify/scrape.ts

@@ -55,3 +55,57 @@ export async function convertPageToMarkdown(url: string): Promise<string> {
   const markdown = await parseMarkdown(cleanedHtml, url);
   return markdown;
 }
+
+/**
+ * Fetches a webpage, strips scripts/styles/images/etc,
+ * returns clean HTML.
+ */
+export async function convertPageToHTML(url: string): Promise<string> {
+  const browser = await chromium.launch();
+  const page = await browser.newPage();
+
+  await page.goto(url, { waitUntil: "networkidle" });
+
+  await page.addInitScript(() => {
+    const selectors = [
+      "script",
+      "style",
+      "link[rel='stylesheet']",
+      "noscript",
+      "meta",
+      "svg",
+      "img",
+      "picture",
+      "source",
+      "video",
+      "audio",
+      "iframe",
+      "object",
+      "embed"
+    ];
+
+    selectors.forEach(sel => {
+      document.querySelectorAll(sel).forEach(e => e.remove());
+    });
+
+    // Remove inline event handlers (onclick, onload…)
+    const all = document.querySelectorAll("*");
+    all.forEach(el => {
+      [...el.attributes].forEach(attr => {
+        if (attr.name.startsWith("on")) {
+          el.removeAttribute(attr.name);
+        }
+      });
+    });
+  });
+
+  // Re-extract HTML after cleanup
+  const cleanedHtml = await page.evaluate(() => {
+    return document.documentElement.outerHTML;
+  });
+
+  await browser.close();
+
+  // Return cleaned HTML directly
+  return cleanedHtml;
+}

server/src/models/Robot.ts

@@ -9,8 +9,9 @@ interface RobotMeta {
   pairs: number;
   updatedAt: string;
   params: any[];
-  type?: 'traditional' | 'markdown';
+  type?: 'extract' | 'scrape';
   url?: string;
+  formats?: ('markdown' | 'html')[];
 }
 
 interface RobotWorkflow {

server/src/pgboss-worker.ts

@@ -20,7 +20,7 @@ import { airtableUpdateTasks, processAirtableUpdates } from './workflow-manageme
 import { io as serverIo } from "./server";
 import { sendWebhook } from './routes/webhook';
 import { BinaryOutputService } from './storage/mino';
-import { convertPageToMarkdown } from './markdownify/scrape';
+import { convertPageToMarkdown, convertPageToHTML } from './markdownify/scrape';
 
 if (!process.env.DB_USER || !process.env.DB_PASSWORD || !process.env.DB_HOST || !process.env.DB_PORT || !process.env.DB_NAME) {
     throw new Error('Failed to start pgboss worker: one or more required environment variables are missing.');
@@ -189,12 +189,14 @@ async function processRunExecution(job: Job<ExecuteRunData>) {
         throw new Error(`Recording for run ${data.runId} not found`);
       }
 
-      if (recording.recording_meta.type === 'markdown') {
+      if (recording.recording_meta.type === 'scrape') {
-        logger.log('info', `Executing markdown robot for run ${data.runId}`);
+        logger.log('info', `Executing scrape robot for run ${data.runId}`);
+
+        const formats = recording.recording_meta.formats || ['markdown'];
 
         await run.update({
           status: 'running',
-          log: 'Converting page to markdown'
+          log: `Converting page to ${formats.join(', ')}`
         });
 
         try {
@@ -204,20 +206,34 @@ async function processRunExecution(job: Job<ExecuteRunData>) {
             throw new Error('No URL specified for markdown robot');
           }
 
-          const markdown = await convertPageToMarkdown(url);
+          let markdown = '';
+          let html = '';
+          const serializableOutput: any = {};
 
+          // Markdown conversion
+          if (formats.includes('markdown')) {
+            markdown = await convertPageToMarkdown(url);
+            serializableOutput.markdown = [{ content: markdown }];
+          }
+
+          // HTML conversion
+          if (formats.includes('html')) {
+            html = await convertPageToHTML(url);
+            serializableOutput.html = [{ content: html }];
+          }
+
+          // Success update
           await run.update({
             status: 'success',
             finishedAt: new Date().toLocaleString(),
-            log: 'Markdown conversion completed successfully',
+            log: `${formats.join(', ').toUpperCase()} conversion completed successfully`,
-            serializableOutput: {
+            serializableOutput,
-              markdown: [{ content: markdown }]
-            },
             binaryOutput: {},
           });
 
           logger.log('info', `Markdown robot execution completed for run ${data.runId}`);
 
+          // Notify sockets
           try {
             const completionData = {
               runId: data.runId,
@@ -233,15 +249,19 @@ async function processRunExecution(job: Job<ExecuteRunData>) {
             logger.log('warn', `Failed to send run-completed notification for markdown robot run ${data.runId}: ${socketError.message}`);
           }
 
+          // Webhooks
           try {
-            const webhookPayload = {
+            const webhookPayload: any = {
               runId: data.runId,
               robotId: plainRun.robotMetaId,
               robotName: recording.recording_meta.name,
               status: 'success',
               finishedAt: new Date().toLocaleString(),
-              markdown: markdown
             };
+
+            if (formats.includes('markdown')) webhookPayload.markdown = markdown;
+            if (formats.includes('html')) webhookPayload.html = html;
+
             await sendWebhook(plainRun.robotMetaId, 'run_completed', webhookPayload);
             logger.log('info', `Webhooks sent successfully for markdown robot run ${data.runId}`);
           } catch (webhookError: any) {
@@ -251,13 +271,14 @@ async function processRunExecution(job: Job<ExecuteRunData>) {
           await destroyRemoteBrowser(browserId, data.userId);
 
           return { success: true };
+
         } catch (error: any) {
-          logger.log('error', `Markdown conversion failed for run ${data.runId}: ${error.message}`);
+          logger.log('error', `${formats.join(', ')} conversion failed for run ${data.runId}: ${error.message}`);
 
           await run.update({
             status: 'failed',
             finishedAt: new Date().toLocaleString(),
-            log: `Markdown conversion failed: ${error.message}`,
+            log: `${formats.join(', ').toUpperCase()} conversion failed: ${error.message}`,
           });
 
           try {

server/src/routes/storage.ts

@@ -440,9 +440,9 @@ router.post('/recordings/:id/duplicate', requireSignIn, async (req: Authenticate
 /**
  * POST endpoint for creating a markdown robot
  */
-router.post('/recordings/markdown', requireSignIn, async (req: AuthenticatedRequest, res) => {
+router.post('/recordings/scrape', requireSignIn, async (req: AuthenticatedRequest, res) => {
   try {
-    const { url, name } = req.body;
+    const { url, name, formats } = req.body;
 
     if (!url) {
       return res.status(400).json({ error: 'The "url" field is required.' });
@@ -459,6 +459,18 @@ router.post('/recordings/markdown', requireSignIn, async (req: AuthenticatedRequ
       return res.status(400).json({ error: 'Invalid URL format' });
     }
 
+    // Validate format
+    const validFormats = ['markdown', 'html'];
+
+    if (!Array.isArray(formats) || formats.length === 0) {
+      return res.status(400).json({ error: 'At least one output format must be selected.' });
+    }
+
+    const invalid = formats.filter(f => !validFormats.includes(f));
+    if (invalid.length > 0) {
+      return res.status(400).json({ error: `Invalid formats: ${invalid.join(', ')}` });
+    }
+
     const robotName = name || `Markdown Robot - ${new URL(url).hostname}`;
     const currentTimestamp = new Date().toLocaleString();
     const robotId = uuid();
@@ -473,8 +485,9 @@ router.post('/recordings/markdown', requireSignIn, async (req: AuthenticatedRequ
         updatedAt: currentTimestamp,
         pairs: 0,
         params: [],
-        type: 'markdown',
+        type: 'scrape',
         url: url,
+        formats: formats,
       },
       recording: { workflow: [] },
       google_sheet_email: null,

server/src/workflow-management/scheduler/index.ts

@@ -15,7 +15,7 @@ import { WorkflowFile } from "maxun-core";
 import { Page } from "playwright";
 import { sendWebhook } from "../../routes/webhook";
 import { airtableUpdateTasks, processAirtableUpdates } from "../integrations/airtable";
-import { convertPageToMarkdown } from "../../markdownify/scrape";
+import { convertPageToMarkdown, convertPageToHTML } from "../../markdownify/scrape";
 chromium.use(stealthPlugin());
 
 async function createWorkflowAndStoreMetadata(id: string, userId: string) {
@@ -208,12 +208,14 @@ async function executeRun(id: string, userId: string) {
       }
     }
 
-    if (recording.recording_meta.type === 'markdown') {
+    if (recording.recording_meta.type === 'scrape') {
-      logger.log('info', `Executing markdown robot for scheduled run ${id}`);
+      logger.log('info', `Executing scrape robot for scheduled run ${id}`);
+
+      const formats = recording.recording_meta.formats || ['markdown'];
 
       await run.update({
         status: 'running',
-        log: 'Converting page to markdown'
+        log: `Converting page to: ${formats.join(', ')}`
       });
 
       try {
@@ -226,9 +228,15 @@ async function executeRun(id: string, userId: string) {
         };
 
         serverIo.of('/queued-run').to(`user-${userId}`).emit('run-started', runStartedData);
-        logger.log('info', `Markdown robot run started notification sent for run: ${plainRun.runId} to user-${userId}`);
+        logger.log(
+          'info',
+          `Markdown robot run started notification sent for run: ${plainRun.runId} to user-${userId}`
+        );
       } catch (socketError: any) {
-        logger.log('warn', `Failed to send run-started notification for markdown robot run ${plainRun.runId}: ${socketError.message}`);
+        logger.log(
+          'warn',
+          `Failed to send run-started notification for markdown robot run ${plainRun.runId}: ${socketError.message}`
+        );
       }
 
       try {
@@ -238,20 +246,33 @@ async function executeRun(id: string, userId: string) {
           throw new Error('No URL specified for markdown robot');
         }
 
-        const markdown = await convertPageToMarkdown(url);
+        let markdown = '';
+        let html = '';
+        const serializableOutput: any = {};
+
+        // Markdown conversion
+        if (formats.includes('markdown')) {
+          markdown = await convertPageToMarkdown(url);
+          serializableOutput.markdown = [{ content: markdown }];
+        }
+
+        // HTML conversion
+        if (formats.includes('html')) {
+          html = await convertPageToHTML(url);
+          serializableOutput.html = [{ content: html }];
+        }
 
         await run.update({
           status: 'success',
           finishedAt: new Date().toLocaleString(),
-          log: 'Markdown conversion completed successfully',
+          log: `${formats.join(', ')} conversion completed successfully`,
-          serializableOutput: {
+          serializableOutput,
-            markdown: [{ content: markdown }]
-          },
           binaryOutput: {},
         });
 
         logger.log('info', `Markdown robot execution completed for scheduled run ${id}`);
 
+        // Run-completed socket notifications
         try {
           const completionData = {
             runId: plainRun.runId,
@@ -264,40 +285,53 @@ async function executeRun(id: string, userId: string) {
           serverIo.of(plainRun.browserId).emit('run-completed', completionData);
           serverIo.of('/queued-run').to(`user-${userId}`).emit('run-completed', completionData);
         } catch (socketError: any) {
-          logger.log('warn', `Failed to send run-completed notification for markdown robot run ${id}: ${socketError.message}`);
+          logger.log(
+            'warn',
+            `Failed to send run-completed notification for markdown robot run ${id}: ${socketError.message}`
+          );
         }
 
-        const webhookPayload = {
+        // Webhook payload
+        const webhookPayload: any = {
           robot_id: plainRun.robotMetaId,
           run_id: plainRun.runId,
           robot_name: recording.recording_meta.name,
           status: 'success',
           started_at: plainRun.startedAt,
           finished_at: new Date().toLocaleString(),
-          markdown: markdown,
           metadata: {
             browser_id: plainRun.browserId,
             user_id: userId,
           }
         };
 
+        if (formats.includes('markdown')) webhookPayload.markdown = markdown;
+        if (formats.includes('html')) webhookPayload.html = html;
+
         try {
           await sendWebhook(plainRun.robotMetaId, 'run_completed', webhookPayload);
-          logger.log('info', `Webhooks sent successfully for markdown robot scheduled run ${plainRun.runId}`);
+          logger.log(
+            'info',
+            `Webhooks sent successfully for markdown robot scheduled run ${plainRun.runId}`
+          );
         } catch (webhookError: any) {
-          logger.log('warn', `Failed to send webhooks for markdown robot run ${plainRun.runId}: ${webhookError.message}`);
+          logger.log(
+            'warn',
+            `Failed to send webhooks for markdown robot run ${plainRun.runId}: ${webhookError.message}`
+          );
         }
 
         await destroyRemoteBrowser(plainRun.browserId, userId);
 
         return true;
+
       } catch (error: any) {
-        logger.log('error', `Markdown conversion failed for scheduled run ${id}: ${error.message}`);
+        logger.log('error', `${formats.join(', ')} conversion failed for scheduled run ${id}: ${error.message}`);
 
         await run.update({
           status: 'failed',
           finishedAt: new Date().toLocaleString(),
-          log: `Markdown conversion failed: ${error.message}`,
+          log: `${formats.join(', ')} conversion failed: ${error.message}`,
         });
 
         try {
@@ -312,7 +346,10 @@ async function executeRun(id: string, userId: string) {
           serverIo.of(plainRun.browserId).emit('run-completed', failureData);
           serverIo.of('/queued-run').to(`user-${userId}`).emit('run-completed', failureData);
         } catch (socketError: any) {
-          logger.log('warn', `Failed to send run-failed notification for markdown robot run ${id}: ${socketError.message}`);
+          logger.log(
+            'warn',
+            `Failed to send run-failed notification for markdown robot run ${id}: ${socketError.message}`
+          );
         }
 
         await destroyRemoteBrowser(plainRun.browserId, userId);

src/components/robot/RecordingsTable.tsx

@@ -110,7 +110,10 @@ const TableRowMemoized = memo(({ row, columns, handlers }: any) => {
             case 'integrate':
               return (
                 <MemoizedTableCell key={column.id} align={column.align}>
-                  <MemoizedIntegrateButton handleIntegrate={() => handlers.handleIntegrateRecording(row.id, row.name, row.params || [])} />
+                  <MemoizedIntegrateButton
+                    handleIntegrate={() => handlers.handleIntegrateRecording(row.id, row.name, row.params || [])}
+                    robotType={row.type}
+                  />
                 </MemoizedTableCell>
               );
             case 'options':
@@ -121,6 +124,7 @@ const TableRowMemoized = memo(({ row, columns, handlers }: any) => {
                     handleEdit={() => handlers.handleEditRobot(row.id, row.name, row.params || [])}
                     handleDuplicate={() => handlers.handleDuplicateRobot(row.id, row.name, row.params || [])}
                     handleDelete={() => handlers.handleDelete(row.id)}
+                    robotType={row.type}
                   />
                 </MemoizedTableCell>
               );
@@ -709,13 +713,22 @@ const ScheduleButton = ({ handleSchedule }: ScheduleButtonProps) => {
 
 interface IntegrateButtonProps {
   handleIntegrate: () => void;
+  robotType: string;
 }
 
-const IntegrateButton = ({ handleIntegrate }: IntegrateButtonProps) => {
+const IntegrateButton = ({ handleIntegrate, robotType }: IntegrateButtonProps) => {
+  const isDisabled = robotType === 'scrape';
+
   return (
-    <IconButton aria-label="add" size="small" onClick={() => {
+    <IconButton
-      handleIntegrate();
+      aria-label="integrate"
-    }}
+      size="small"
+      onClick={isDisabled ? undefined : handleIntegrate}
+      disabled={isDisabled}
+      sx={{
+        opacity: isDisabled ? 0.4 : 1,
+        cursor: isDisabled ? 'not-allowed' : 'pointer',
+      }}
     >
       <Power />
     </IconButton>
@@ -742,9 +755,10 @@ interface OptionsButtonProps {
   handleEdit: () => void;
   handleDelete: () => void;
   handleDuplicate: () => void;
+  robotType: string;
 }
 
-const OptionsButton = ({ handleRetrain, handleEdit, handleDelete, handleDuplicate }: OptionsButtonProps) => {
+const OptionsButton = ({ handleRetrain, handleEdit, handleDelete, handleDuplicate, robotType }: OptionsButtonProps) => {
   const [anchorEl, setAnchorEl] = React.useState<null | HTMLElement>(null);
 
   const handleClick = (event: React.MouseEvent<HTMLElement>) => {
@@ -771,34 +785,33 @@ const OptionsButton = ({ handleRetrain, handleEdit, handleDelete, handleDuplicat
         open={Boolean(anchorEl)}
         onClose={handleClose}
       >
-        <MenuItem onClick={() => { handleRetrain(); handleClose(); }}>
+        {robotType !== 'scrape' && (
-          <ListItemIcon>
+          <MenuItem onClick={() => { handleRetrain(); handleClose(); }}>
-            <Refresh fontSize="small" />
+            <ListItemIcon>
-          </ListItemIcon>
+              <Refresh fontSize="small" />
-          <ListItemText>{t('recordingtable.retrain')}</ListItemText>
+            </ListItemIcon>
-        </MenuItem>
+            <ListItemText>Retrain</ListItemText>
+          </MenuItem>
+        )}
 
         <MenuItem onClick={() => { handleEdit(); handleClose(); }}>
-          <ListItemIcon>
+          <ListItemIcon><Edit fontSize="small" /></ListItemIcon>
-            <Edit fontSize="small" />
+          <ListItemText>Edit</ListItemText>
-          </ListItemIcon>
-          <ListItemText>{t('recordingtable.edit')}</ListItemText>
         </MenuItem>
 
         <MenuItem onClick={() => { handleDelete(); handleClose(); }}>
-          <ListItemIcon>
+          <ListItemIcon><DeleteForever fontSize="small" /></ListItemIcon>
-            <DeleteForever fontSize="small" />
+          <ListItemText>Delete</ListItemText>
-          </ListItemIcon>
-          <ListItemText>{t('recordingtable.delete')}</ListItemText>
         </MenuItem>
 
-        <MenuItem onClick={() => { handleDuplicate(); handleClose(); }}>
+        {robotType !== 'scrape' && (
-          <ListItemIcon>
+          <MenuItem onClick={() => { handleDuplicate(); handleClose(); }}>
-            <ContentCopy fontSize="small" />
+            <ListItemIcon><ContentCopy fontSize="small" /></ListItemIcon>
-          </ListItemIcon>
+            <ListItemText>Duplicate</ListItemText>
-          <ListItemText>{t('recordingtable.duplicate')}</ListItemText>
+          </MenuItem>
-        </MenuItem>
+        )}
       </Menu>
+
     </>
   );
 };

src/components/robot/pages/RobotCreate.tsx

@@ -15,12 +15,16 @@ import {
   Container,
   CardContent,
   Tabs,
-  Tab
+  Tab,
+  RadioGroup,
+  Radio,
+  FormControl,
+  FormLabel
 } from '@mui/material';
 import { ArrowBack, PlayCircleOutline, Article, Code, Description } from '@mui/icons-material';
 import { useGlobalInfoStore } from '../../../context/globalInfo';
 import { canCreateBrowserInState, getActiveBrowserId, stopRecording } from '../../../api/recording';
-import { createMarkdownRobot } from "../../../api/storage";
+import { createScrapeRobot } from "../../../api/storage";
 import { AuthContext } from '../../../context/auth';
 import { GenericModal } from '../../ui/GenericModal';
 
@@ -54,11 +58,12 @@ const RobotCreate: React.FC = () => {
 
   const [tabValue, setTabValue] = useState(0);
   const [url, setUrl] = useState('');
-  const [markdownRobotName, setMarkdownRobotName] = useState('');
+  const [scrapeRobotName, setScrapeRobotName] = useState('');
   const [needsLogin, setNeedsLogin] = useState(false);
   const [isLoading, setIsLoading] = useState(false);
   const [isWarningModalOpen, setWarningModalOpen] = useState(false);
   const [activeBrowserId, setActiveBrowserId] = useState('');
+  const [outputFormats, setOutputFormats] = useState<string[]>([]);
 
   const { state } = React.useContext(AuthContext);
   const { user } = state;
@@ -200,7 +205,7 @@ const RobotCreate: React.FC = () => {
             }}
           >
             <Tab label="Extract" id="extract-robot" aria-controls="extract-robot" />
-            <Tab label="Markdown" id="markdown-robot" aria-controls="markdown-robot" />
+            <Tab label="Scrape" id="scrape-robot" aria-controls="scrape-robot" />
           </Tabs>
         </Box>
 
@@ -370,7 +375,7 @@ const RobotCreate: React.FC = () => {
               />
 
               <Typography variant="body2" color="text.secondary" mb={3}>
-                Turn websites into LLM-ready Markdown content for AI apps.
+                Turn websites into LLM-ready Markdown or clean HTML content for AI apps.
               </Typography>
 
               <Box sx={{ width: '100%', maxWidth: 700, mb: 2 }}>
@@ -378,8 +383,8 @@ const RobotCreate: React.FC = () => {
                   placeholder="Example: YC Companies Scraper"
                   variant="outlined"
                   fullWidth
-                  value={markdownRobotName}
+                  value={scrapeRobotName}
-                  onChange={(e) => setMarkdownRobotName(e.target.value)}
+                  onChange={(e) => setScrapeRobotName(e.target.value)}
                   sx={{ mb: 2 }}
                   label="Robot Name"
                 />
@@ -390,7 +395,44 @@ const RobotCreate: React.FC = () => {
                   value={url}
                   onChange={(e) => setUrl(e.target.value)}
                   label="Website URL"
+                  sx={{ mb: 2 }}
                 />
+
+                <FormControl component="fieldset" sx={{ width: '100%', textAlign: 'left' }}>
+                  <FormLabel component="legend" sx={{ mb: 1 }}>Output Format (Select at least one)</FormLabel>
+
+                  <FormControlLabel
+                    control={
+                      <Checkbox
+                        checked={outputFormats.includes('markdown')}
+                        onChange={(e) => {
+                          if (e.target.checked) {
+                            setOutputFormats([...outputFormats, 'markdown']);
+                          } else {
+                            setOutputFormats(outputFormats.filter(f => f !== 'markdown'));
+                          }
+                        }}
+                      />
+                    }
+                    label="Markdown"
+                  />
+
+                  <FormControlLabel
+                    control={
+                      <Checkbox
+                        checked={outputFormats.includes('html')}
+                        onChange={(e) => {
+                          if (e.target.checked) {
+                            setOutputFormats([...outputFormats, 'html']);
+                          } else {
+                            setOutputFormats(outputFormats.filter(f => f !== 'html'));
+                          }
+                        }}
+                      />
+                    }
+                    label="HTML"
+                  />
+                </FormControl>
               </Box>
 
               <Button
@@ -401,23 +443,28 @@ const RobotCreate: React.FC = () => {
                     notify('error', 'Please enter a valid URL');
                     return;
                   }
-                  if (!markdownRobotName.trim()) {
+                  if (!scrapeRobotName.trim()) {
                     notify('error', 'Please enter a robot name');
                     return;
                   }
+                  if (outputFormats.length === 0) {
+                    notify('error', 'Please select at least one output format');
+                    return;
+                  }
+
                   setIsLoading(true);
-                  const result = await createMarkdownRobot(url, markdownRobotName);
+                  const result = await createScrapeRobot(url, scrapeRobotName, outputFormats);
                   setIsLoading(false);
 
                   if (result) {
                     setRerenderRobots(true);
-                    notify('success', `${markdownRobotName} created successfully!`);
+                    notify('success', `${scrapeRobotName} created successfully!`);
                     navigate('/robots');
                   } else {
                     notify('error', 'Failed to create markdown robot');
                   }
                 }}
-                disabled={!url.trim() || !markdownRobotName.trim() || isLoading}
+                disabled={!url.trim() || !scrapeRobotName.trim() || outputFormats.length === 0 || isLoading}
                 sx={{
                   bgcolor: '#ff00c3',
                   py: 1.4,
@@ -428,7 +475,10 @@ const RobotCreate: React.FC = () => {
                 }}
                 startIcon={isLoading ? <CircularProgress size={20} color="inherit" /> : null}
               >
-                {isLoading ? 'Turning...' : 'Turn to Markdown'}
+                {isLoading
+                  ? "Creating..."
+                  : `Create ${outputFormats.join(" + ").toUpperCase()} Robot`
+                }
               </Button>
             </Box>
           </Card>

src/components/robot/pages/RobotDuplicatePage.tsx

@@ -24,8 +24,9 @@ interface RobotMeta {
   pairs: number;
   updatedAt: string;
   params: any[];
-  type?: 'traditional' | 'markdown';
+  type?: 'extract' | 'scrape';
   url?: string;
+  formats?: ('markdown' | 'html')[];
 }
 
 interface RobotWorkflow {

src/components/robot/pages/RobotEditPage.tsx

@@ -24,8 +24,9 @@ interface RobotMeta {
   pairs: number;
   updatedAt: string;
   params: any[];
-  type?: 'traditional' | 'markdown';
+  type?: 'extract' | 'scrape';
   url?: string;
+  formats?: ('markdown' | 'html')[];
 }
 
 interface RobotWorkflow {

src/components/robot/pages/RobotSettingsPage.tsx

@@ -16,8 +16,9 @@ interface RobotMeta {
   pairs: number;
   updatedAt: string;
   params: any[];
-  type?: 'traditional' | 'markdown';
+  type?: 'extract' | 'scrape';
   url?: string;
+  formats?: ('markdown' | 'html')[];
 }
 
 interface RobotWorkflow {

src/context/globalInfo.tsx

@@ -27,8 +27,9 @@ interface RobotMeta {
     pairs: number;
     updatedAt: string;
     params: any[];
-    type?: 'traditional' | 'markdown';
+    type?: 'extract' | 'scrape';
     url?: string;
+    formats?: ('markdown' | 'html')[];
 }
 
 interface RobotWorkflow {