Step 2: Connect the RSS Source and Filter Recent Items

Pull items from the RSS feed and filter down to only recent entries.

1. In Retrieve RSS Entries, set URL to URL_FEED.
2. In Validate Recent Items, configure a date filter where Left Value is {{ $json.pubDate }} and Right Value is {{ $today.minus({ days: 1 }) }}, using the after operator.
3. Connect Retrieve RSS Entries → Validate Recent Items.

Step 3: Prepare and Limit Articles for Processing

Normalize the article link, cap the total items, and iterate through them for content extraction.

1. In Assign Article Link, add a field named link with value {{ $json.link }}.
2. In Cap Items, set Max Items to 3.
3. Connect Validate Recent Items → Assign Article Link → Cap Items → Iterate Articles.
4. Ensure Iterate Articles has default batch settings so it processes items one by one.

Step 4: Extract and Aggregate Article Content

Scrape each article into markdown, extract key details, and aggregate them into a single dataset.

1. In Markdownify Web Page, set Resource to markdownify and Website URL to {{ $json.link }}.
2. Credential Required: Connect your scrapegraphAIApi credentials in Markdownify Web Page.
3. In Extract Key Details, set Text to {{ $json.result }}, and keep the attribute content description as provided.
4. Ensure Gemini Chat Engine is connected as the language model for Extract Key Details. Credential Required: Connect your googlePalmApi credentials in Gemini Chat Engine.
5. In Aggregate Content, set Field to Aggregate from output.content to Output Field Name text.
6. Connect Iterate Articles → Markdownify Web Page → Extract Key Details → Iterate Articles (loop) → Aggregate Content.

Step 5: Compose the Digest with AI and Structured Output

Summarize aggregated content into a formatted email-ready digest using chained AI models and structured output parsing.

1. In Compose Digest, set Text to {{ JSON.stringify($json.text) }} and keep Prompt Type as define.
2. In Compose Digest, ensure Has Output Parser is enabled (true).
3. Verify OpenAI Summary Model is connected as the language model for Compose Digest. Credential Required: Connect your openAiApi credentials in OpenAI Summary Model.
4. In Structured Output Reader, keep Auto Fix enabled and ensure the schema includes subject and body exactly as defined.
5. Verify OpenAI Format Model is connected as the language model for Structured Output Reader. Credential Required: Connect your openAiApi credentials in OpenAI Format Model. Add credentials to the parent model node, not the parser node.
6. Connect Aggregate Content → Compose Digest and attach Structured Output Reader to Compose Digest as the output parser.

Step 6: Configure Output Delivery (Email and Telegram)

Send the digest to email and Telegram in parallel once the summary is ready.

1. In Dispatch Email Alert, set Send To to [YOUR_EMAIL], Subject to {{ $json.output.subject }}, and Message to {{ $json.output.body }}.
2. Credential Required: Connect your gmailOAuth2 credentials in Dispatch Email Alert.
3. In Post Telegram Update, set Chat ID to [YOUR_ID] and Text to {{ $json.output.body }}.
4. Credential Required: Connect your telegramApi credentials in Post Telegram Update.
5. Compose Digest outputs to both Dispatch Email Alert and Post Telegram Update in parallel.

Step 7: Test & Activate Your Workflow

Run a manual test to validate content extraction and delivery, then enable the workflow for production use.

1. Click Execute Workflow to run Manual Run Trigger and verify items flow through Retrieve RSS Entries and Validate Recent Items.
2. Confirm that Compose Digest outputs a structured subject and body, and that both Dispatch Email Alert and Post Telegram Update receive the same content.
3. Check your inbox and Telegram chat for the digest; successful execution shows a formatted summary message delivered to both channels.
4. Toggle the workflow to Active to enable production use once testing is successful.