AI Employee Feedback Sentiment Analyzer

1. Get your Claude API key from console.anthropic.com and store it as a credential. 2. Create a Typeform survey with anonymous feedback questions: overall satisfaction (1-5), what's going well (open text), what could improve (open text), and any urgent concerns (open text). 3. Create a Google Sheet with columns: Submission Date, Satisfaction Score, Positive Feedback, Improvement Areas, Urgent Concerns, Sentiment (positive/neutral/negative), Key Themes (comma-separated), Urgency Level (low/medium/high/critical), and AI Summary. 4. Set up a Typeform trigger that fires on each new submission. Map the responses to the raw feedback columns in Google Sheets. 5. Add an HTTP Request node to call the Claude API (POST https://api.anthropic.com/v1/messages) with model "claude-sonnet-4-20250514". Send all three open-text responses and the satisfaction score. Ask Claude to return JSON with: sentiment (positive/neutral/negative), themes (array of strings), urgencyLevel (low/medium/high/critical), summary (2-3 sentences), and actionItems (array of strings). 6. Parse the Claude response and update the same Google Sheets row with the analysis columns. 7. Add a Slack node that posts a formatted summary to your HR or leadership channel. Include the sentiment, key themes, urgency level, and any action items. Do not include identifiable information — keep it anonymous. 8. For batch analysis, add a scheduled trigger (e.g., weekly) that reads all new rows, sends them to Claude as a batch, and posts a weekly trends report to Slack.