Step 2: Connect the Scrape Request & Task Status Checks

These nodes start the BrowserAct workflow task and poll for completion.

1. Open Start Scrape Request and set URL to https://api.browseract.com/v2/workflow/run-task.
2. Set Method to POST, enable Send Body, and set Authentication to genericCredentialType with Generic Auth Type httpBearerAuth.
3. Under Body Parameters, set workflow_id to [YOUR_ID].
4. Credential Required: Connect your httpBearerAuth credentials in Start Scrape Request.
5. Open Check Scrape Status and set URL to https://api.browseract.com/v2/workflow/get-task with Send Query enabled.
6. Set the query parameter task_id value to {{ $json.id }}.
7. Credential Required: Connect your httpBearerAuth credentials in Check Scrape Status.

Step 3: Configure Polling Logic and Retry Delays

These nodes validate the task response and loop until the scrape is complete.

1. In Validate Task Response, verify the conditions use {{ $json.error }} (not exists) and {{ $json.id }} (not equals null).
2. Confirm the flow: Validate Task Response → Check Scrape Status on success, and Validate Task Response → Short Delay Loop on failure.
3. Set Short Delay Loop to wait 1 minute and route back to Start Scrape Request.
4. In Check Task Complete, ensure conditions use {{ $json.error }} (not exists) and {{ $json.status }} equals finished.
5. Set Pause Before Retry to wait 2 minutes and route back to Check Scrape Status.

Step 4: Set Up the AI Matching Agent

The AI agent compares your resume against scraped job data and outputs structured results.

1. Open Resume Match Agent and confirm the prompt includes {{ $json.output.string }} and outputs the required format_final_json_response structure.
2. Ensure Resume Match Agent has Needs Fallback enabled and Has Output Parser enabled.
3. Connect Gemini Primary Model to Resume Match Agent as the main language model.
4. Credential Required: Connect your googlePalmApi credentials in Gemini Primary Model.
5. Connect Gemini Fallback Model as the fallback language model and set Model Name to models/gemini-2.0-flash.
6. Credential Required: Connect your googlePalmApi credentials in Gemini Fallback Model.
7. Attach Parse Structured Output as the output parser and keep the JSON Schema Example exactly as defined.

Step 5: Format Results and Send Telegram Alerts

Format the AI output into individual items and send a Telegram message for each match.

1. In Format Result List, keep the JavaScript code that maps $input.first().json.output.Request into individual text objects.
2. Confirm the flow: Resume Match Agent → Format Result List → Dispatch Telegram Alert.
3. Open Dispatch Telegram Alert and set Text to Title: {{ $json.text.Title }} Summary :{{ $json.text.Summary }} Needs: {{ $json.text.Needs }} Educarion: {{ $json.text.Education }} {{ $json.text.Link }}.
4. Set Chat ID to [YOUR_ID].
5. Credential Required: Connect your telegramApi credentials in Dispatch Telegram Alert.

Step 6: Test and Activate Your Workflow

Run a manual test to confirm the scrape, AI matching, formatting, and Telegram notification outputs.

1. Click Execute Workflow on Manual Launch Trigger to run the full flow.
2. Verify Start Scrape Request returns a task ID and Check Scrape Status eventually shows finished.
3. Confirm Resume Match Agent outputs the structured Request array and Format Result List splits items correctly.
4. Check Telegram for messages sent by Dispatch Telegram Alert with title, summary, needs, education, and link.
5. When satisfied, set the workflow to Active for production use.