Step 2: Configure Apify Data Retrieval

Use Apify’s API to run a post search and fetch the latest dataset items.

1. Open Apify Post Search Request and set URL to https://api.apify.com/v2/acts/scraper_one~x-posts-search/run-sync?token=[CONFIGURE_YOUR_TOKEN].
2. Set Method to POST and JSON Body to { "query": "LaraconIn", "resultsCount": 5, "searchType": "latest", "lang":"en" }.
3. Ensure Send Body and Send Headers are enabled, and set the Content-Type header to application/json.
4. Open Fetch Apify Results and set URL to https://api.apify.com/v2/acts/scraper_one~x-posts-search/runs/last/dataset/items?token=[CONFIGURE_YOUR_TOKEN].
5. Connect Apify Post Search Request → Fetch Apify Results → Map Loop Fields.

Step 3: Prepare Records and Check for Duplicates

Normalize the Apify results, loop through each record, and skip already-processed posts.

1. In Map Loop Fields, map fields to:
2. Set id to {{ $json.postId }}, tweet_url to {{ $json.postUrl }}, and tweet_text to {{ $json.postText }}.
3. Connect Map Loop Fields → Iterate Records to process one record at a time.
4. Open Retrieve Post Sheet and set the filter to match ID with {{ $json.id }}.
5. Credential Required: Connect your googleSheetsOAuth2Api credentials in Retrieve Post Sheet.
6. In Duplicate Check, keep the condition {{ $json.ID }} with the exists operator.

Step 4: Set Up AI Sentiment Analysis and Routing

Analyze sentiment for each tweet and route to the appropriate action path.

1. Open Analyze Sentiment and set Text to {{ $('Iterate Records').item.json.tweet_text }}.
2. Confirm Prompt Type is define and Has Output Parser is enabled.
3. Open Parse Structured Output and keep the JSON schema example { "category": "neutral" }.
4. Open Chat Model Engine and select model gpt-4o-mini.
5. Credential Required: Connect your openAiApi credentials in Chat Model Engine.
6. Ensure Chat Model Engine is connected as the language model for Analyze Sentiment, and Parse Structured Output is connected as the output parser.
7. Open Route by Sentiment and verify the rules compare {{ $json.output.category }} to Positive, Negative, and Neutral.

Step 5: Configure Reply Generation and Notifications

Generate replies for positive sentiment and send alerts for negative posts, then update the Google Sheet.

1. Open Compose Reply AI and ensure the user message uses {{ $('Map Loop Fields').item.json.tweet_text }}.
2. Credential Required: Connect your openAiApi credentials in Compose Reply AI.
3. Open Slack Alert for Negative and set Text to Received a Negative retweet on {{ $('Duplicate Check').item.json.TweetUrl }}. Consider if a response or clarification is needed..
4. Credential Required: Connect your slackOAuth2Api credentials in Slack Alert for Negative.
5. Open Update Post Sheet and confirm Operation is appendOrUpdate.
6. Map columns in Update Post Sheet to: ID {{ $('Map Loop Fields').item.json.id }}, TweetUrl {{ $('Map Loop Fields').item.json.tweet_url }}, TweetText {{ $('Map Loop Fields').item.json.tweet_text }}, Post Reply {{ $json.message.content }}, and Sentiment {{ $('Route by Sentiment').item.json.output.category }}.
7. Credential Required: Connect your googleSheetsOAuth2Api credentials in Update Post Sheet.

Step 6: Test and Activate Your Workflow

Run a full test to confirm Apify retrieval, sentiment analysis, routing, and sheet updates are working as expected.

1. Click Execute Workflow to run the workflow manually.
2. Verify Fetch Apify Results returns new posts and Map Loop Fields creates id, tweet_url, and tweet_text.
3. Check that Route by Sentiment directs items correctly and that Slack Alert for Negative sends a message for negative posts.
4. Confirm Update Post Sheet writes or updates rows with reply and sentiment values.
5. When satisfied, toggle the workflow Active to enable scheduled runs.