Xiaohongshu Insight

Provides daily insights from 2000+ Xiaohongshu viral posts to support content creation, trend analysis, and creative inspiration based on engagement data.

Security Scan

The skill claims continuous, platform-wide collection of 2,000+ Xiaohongshu viral posts daily, but the shipped files contain only mock-data scripts and no network/collection logic, credentials, or install steps — the implementation does not match the stated capability.

Purpose & Capability

The manifest and SKILL.md tout continuous daily collection of 2000+ viral posts and realtime/hourly updates, but the included scripts operate on hardcoded/mock data and contain no network calls, API clients, scraping logic, or dependencies for data ingestion. No homepage or source URL is provided to justify where live data would come from.

Instruction Scope

Runtime instructions describe querying and analysis of viral posts and list update frequencies, but the SKILL.md and scripts only instruct running local scripts that read/generate mock datasets. The instructions claim external data collection/update schedules without any procedural steps or endpoints to perform those actions.

Install Mechanism

No install spec or external downloads are present (instruction-only with local Python scripts). This is low-risk from an install/execution-vector perspective.

Credentials

No environment variables, credentials, or config paths are requested, yet the skill's supposed purpose (platform-wide data collection) would normally require API keys, auth tokens, or access to a data pipeline. The absence of any declared credentials is inconsistent with the claimed capability.

Persistence & Privilege

The skill does not request always:true, has no install-time persistence, and does not modify other skill/system configs. Autonomous invocation is allowed by default but not combined with other red flags here.

Scan Findings in Context

[no_pre_scan_findings] unexpected: Static scanner found no injection signals, but that doesn't validate the capability — absence of network code and absence of credentials is the main coherence issue.

What to consider before installing

This package appears to be a demo or template rather than a working collector: the scripts only use generated/mock data and there is no code to fetch real Xiaohongshu posts, no API endpoints, no install steps, and no credentials requested. Before installing or relying on this skill, ask the publisher for: (1) the source repository or homepage; (2) the actual data ingestion code or API endpoints used to collect posts; (3) what credentials (if any) are required and why; and (4) confirmation that collection methods comply with Xiaohongshu's Terms of Service and privacy laws. If you plan to run the bundled scripts, treat them as safe-to-inspect local Python demos but do not assume they provide live data. If the skill is later extended to perform network access or scraping, review those new files carefully for hidden endpoints, credential use, or exfiltration; run any network-capable version in a sandbox and avoid providing unrelated credentials.

Xiaohongshu Data Insights

Overview

A professional data insight tool built specifically for Xiaohongshu content creation. Continuously collects over 2000+ viral posts daily across the platform, empowering creators with data references, traffic trend insights, and creative inspiration.

Use Cases

Use when users mention Xiaohongshu viral posts, post data, traffic trends, creative inspiration, low-follower viral posts, interaction spikes, etc.

Core Capabilities

1. Viral Post Collection

Automatically collects 2000+ viral posts daily based on the following criteria:

• Low-Follower Viral Posts: Follower count < 5000, yet post interactions exceed the category average by more than 3x
• Periodic High-Engagement: Like growth > 1000 within 7 days, with a clear sustained growth trend
• Single-Day Interaction Spike: Daily increase in likes + saves + comments > 500
• Sustained Interaction Growth: Interactions increase for 3 consecutive days, with average daily growth rate > 20%

2. Data Dimensions

Each viral post includes the following data:

Dimension	Description
Basic Information	Title, cover image, author, publish time, category tags
Interaction Data	Likes, saves, comments, shares, and growth rates
Author Profile	Follower count, total posts, average interactions, category distribution
Content Features	Title keywords, cover style, content type, word count range
Traffic Curve	Interaction data at 24h/48h/72h/7d intervals after publishing

3. Use Scenarios

Scenario 1: Finding Creative Inspiration

User: What are some recent low-follower viral posts in the beauty category?
→ Filter category=Beauty, follower count<5000, sort by interaction volume
→ Return TOP 20 viral posts with key feature analysis

Scenario 2: Analyzing Traffic Trends

User: What common characteristics do this week's viral fashion posts share?
→ Aggregate analysis of viral posts in Fashion category over the past 7 days
→ Extract title keywords, cover features, publishing time distribution
→ Output trend insight report

Scenario 3: Competitor Account Research

User: Why does this blogger consistently produce viral content?
→ Analyze the blogger's historical viral posts
→ Compare with other bloggers in the same category
→ Output success factor analysis

Quick Start

Query Viral Posts

Use scripts/query_notes.py to query viral post data:

python scripts/query_notes.py --category Beauty --days 7 --limit 50

Parameter descriptions:

• --category: Category (Beauty/Fashion/Food/Travel/Home/Parenting/Career/Emotions/Knowledge/Entertainment)
• --days: Number of days to query (default 7 days)
• --limit: Number of results to return (default 20)
• --sort: Sort field (engagement/growth/fans)
• --min_engagement: Minimum interaction threshold

Analyze Trend Features

Use scripts/analyze_trends.py to analyze category trends:

python scripts/analyze_trends.py --category Fashion --output report.md

Output includes:

• TOP 20 Trending Keywords
• High-Frequency Cover Styles
• Optimal Publishing Time Windows
• Viral Title Formulas
• Engagement Rate Distribution

Export Data

Use scripts/export_data.py to export data:

python scripts/export_data.py --format xlsx --output viral_data.xlsx

Supported formats: xlsx, csv, json

Data Field Descriptions

See references/data_schema.md for details.

Viral Post Detection Algorithm

See references/viral_criteria.md for details.

Notes

1. Data is sourced from publicly available posts and is for creative reference only
2. It is recommended to selectively draw inspiration based on your own account positioning
3. Avoid directly copying content; focus on creative transformation
4. Data is updated daily; pay attention to timeliness