Step 2: Connect Apify to Fetch TikTok Data

This step runs the Apify actor to scrape TikTok videos matching a rental hashtag.

1. Add the Run Actor and Fetch Dataset node.
2. Set Operation to Run actor and get dataset.
3. Set Memory to 8192.
4. Set Actor ID to your Apify actor (value shows as [YOUR_ID] in the template).
5. Set Custom Body to the full JSON expression:
   ={ "excludePinnedPosts": false, "hashtags": [ "賃貸" ], "maxProfilesPerQuery": 1, "proxyCountryCode": "None", "resultsPerPage": 1, "scrapeRelatedVideos": false, "shouldDownloadAvatars": false, "shouldDownloadCovers": false, "shouldDownloadMusicCovers": false, "shouldDownloadSlideshowImages": false, "shouldDownloadSubtitles": false, "shouldDownloadVideos": false }
6. Credential Required: Connect your Apify OAuth2 API credentials.
7. Connect Run Actor and Fetch Dataset to AI Reasoning Agent.

Step 3: Set Up the AI Reasoning and Summary Agents

Two AI agents select a suitable property video and then generate a short reason for recommendation.

1. Add the AI Reasoning Agent node and set Text to:
   =20代カップル向けの物件を動画内から選択して下さい 説明や前置きは省いて下さい URLのみを出力して下さい --- #URL {{ $json.webVideoUrl }}
2. Ensure OpenRouter Chat Engine is connected as the language model for AI Reasoning Agent.
3. Credential Required: Connect your OpenRouter credentials in OpenRouter Chat Engine (credentials are set on the language model node, not the agent).
4. Add the AI Summary Agent node and set Text to:
   =URLの動画を見て、20代カップルにおすすめな理由を出力して下さい。 説明や、指示文などは入れないで下さい URLと説明で分けて下さい {{ $json.output }}
5. Ensure OpenRouter Chat Engine 2 is connected as the language model for AI Summary Agent.
6. Credential Required: Connect your OpenRouter credentials in OpenRouter Chat Engine 2 (credentials are set on the language model node, not the agent).
7. Connect AI Reasoning Agent → AI Summary Agent.

Step 4: Configure Output Destinations (Google Sheets and Slack)

The summary is stored in Google Sheets and then posted to Slack.

1. Add the Update Sheet Row node.
2. Set Operation to appendOrUpdate.
3. Select your spreadsheet in Document ID (template shows [YOUR_ID]).
4. Select Sheet Name (template shows gid=0 / シート1).
5. Leave the column mapping in Columns set to autoMapInputData so the output field is captured.
6. Credential Required: Connect your Google Sheets credentials.
7. Add the Post Slack Notification node and set Text to:
   =おすすめの物件情報です {{ $json.output }}
8. Set Select to channel and choose your Channel ID (template shows [YOUR_ID]).
9. Credential Required: Connect your Slack OAuth2 credentials.
10. Connect AI Summary Agent → Update Sheet Row → Post Slack Notification.

Step 5: Test and Activate Your Workflow

Run a manual test to confirm the TikTok scrape, AI analysis, sheet update, and Slack post all succeed.

1. Click Execute Workflow and trigger Manual Launch Trigger.
2. Verify Run Actor and Fetch Dataset returns TikTok video data with webVideoUrl.
3. Confirm AI Reasoning Agent outputs a single URL and AI Summary Agent returns URL + explanation.
4. Check the Google Sheet row was appended/updated by Update Sheet Row.
5. Confirm a message was posted in Slack by Post Slack Notification.
6. When satisfied, switch the workflow to Active for production use.