Sentiment Monitoring

This handles brand sentiment tracking on Xiaohongshu with a three-level framework: monitoring mentions and comments in real time, analyzing sentiment patterns and trends, and coordinating responses before issues spiral. You get a classification system for positive, neutral, negative, and crisis-level sentiment (with Chinese indicators like "太棒了" vs "踩雷"), automated alert triggers when sentiment drops more than 10% in 24 hours, and a daily monitoring routine with morning, midday, and evening checks. The baseline setup is solid, measuring at least 500 comments to establish your sentiment percentage benchmarks. Best for brands running active campaigns or managing reputation risks. The monitoring infrastructure covers everything from native Xiaohongshu analytics to third-party tools like Huitun and Chanmama, plus manual tracking when you need granular control.

skills.sh installs: 263
GitHub stars: 539
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add vivy-yi/xiaohongshu-skills --skill sentiment-monitoring --agent claude-code

Installs into .claude/skills of the current project.

Files

Sentiment Monitoring (舆情监控)

Overview

Sentiment monitoring is the systematic tracking and analysis of brand mentions, comments, and conversations on Xiaohongshu to understand public perception, identify reputation risks early, and respond proactively to maintain positive brand image.

When to Use

Use when:

Monitoring brand health and public perception
Tracking sentiment during marketing campaigns
Identifying and addressing negative feedback early
Managing PR crises or reputation issues
Analyzing competitor sentiment and positioning
Measuring brand sentiment changes over time
Responding to customer complaints or concerns

Do NOT use when:

Monitoring basic engagement metrics (use data-analytics)
Tracking daily content performance (use data-analytics)
Analyzing competitor content strategies (use competitive-analysis)

Core Pattern

Before (blind to public opinion):

❌ "No idea what people are saying about our brand"
❌ "Negative comment goes viral before we notice"
❌ "Can't measure if our campaigns improved perception"
❌ "Reputation crisis catches us completely off guard"

SKILL.md on GitHub