Discover dark horse viral posts that are quietly gaining traction on Xiaohongshu, and see which accounts and content have suddenly taken off recently.
Trigger Scenarios
Use this Skill when users make the following requests:
- "Check the Xiaohongshu Dark Horse Viral Post Ranking"
- "Dark horse notes in the XX category" (e.g., "Home Renovation Dark Horse Viral Post Ranking")
- "See what dark horse content has appeared recently on Xiaohongshu"
- "Check the dark horse viral post ranking for a certain day"
If the category entered by the user cannot be matched to a standard category, output under "All Categories"; when a query returns no results, state this truthfully and do not automatically query other categories.
Data Description
Ranking Criteria
- Ranking Positioning: Dark Horse Viral Post Ranking, presenting dark horse notes with outstanding recent interaction performance,not limited by follower tier; ranked daily (returned notes' publication dates are concentrated on the day of the statistical window or the previous day),notthe seven-day viral note weekly ranking.
- Category Filtering: supports 25 content categories; defaults to "All Categories" when unspecified.
- Date Query: supports querying the ranking for any day within the past 30 days,
rankDateis the statistical windowstart date; when unspecified, defaults to querying the most recent updated data day. - Sorting: sorted by total interaction from high to low, displaying TOP50 (first output TOP20; "View More" shows TOP21-50).
- Data Update Time: updates yesterday's data starting at 17:30 every day (querying yesterday's data before 17:30 will not yet have been generated).
Each note includes
title, author, follower count, publication time, like count, comment count, favorite count, share count, and total interactions.
Supported query categories (25 in total)
All Categories (default), Transportation, Healthcare, Leisure Hobbies, General Miscellaneous, Weddings, Home Renovation, Film and Television Entertainment, Astrology and Relationships, Photography and Recording, Education and Study, Travel and Vacation, Parenting, Daily Life, Scientific Exploration, Digital Technology, Fashion and Outfits, Makeup and Beauty, Personal Care, Delicious Food, Career Development, Pets, News and Information, Sports and Fitness, Trendy Shoes and Bags.
One-Click Installation
Prerequisites
- Python 3.8+ (uses only the standard library; no additional dependencies need to be installed)
- Honghu Data API Key (go to Honghu hub to register and obtain)
Installation Steps
- Place the skill folder into your Skills directory
- Configure the environment variable
REDFOX_API_KEY(for configuration details, seereferences/core_workflow.md） - No additional dependencies need to be installed; use directly
Workflow
The core execution flow (intent recognition, classification and date handling, script invocation, ranking and pattern analysis output rules, HTML file package generation and automatic display, empty results and error handling, etc.) is detailed in
references/core_workflow.md. The Agent must follow all rules in that file during execution. For the complete data interface and field descriptions, seereferences/api-spec.md。
Usage Guide
Describe your query needs directly in natural language, no need to memorize commands.
| Intent | Example Phrasing | Effect |
|---|---|---|
| Query the default ranking | "Show me the Xiaohongshu dark horse viral post ranking" | Output the comprehensive dark horse viral post ranking (default most recent data date) |
| Specify a category | "Check the makeup and beauty dark horse viral post ranking" | Output only that category's ranking |
| Specify a date | "Check the dark horse ranking for 2026-09-20" | Query by the specified statistics window start date |
| View more | "View more" | Output the remaining ranking from TOP21-50 |
| Subscribe to push notifications | "Subscribe" or reply "1" | Automatically push the latest dark horse viral post ranking TOP50 every day at 19:30 |
| Export file package | Reply "2" | Generate a complete Xiaohongshu-style page, supports PDF export |
Output Content
After a successful query, the results are output completely in order:
- Ranking table: serial number, note information (title and author homepage are both clickable), total interactions, likes, comments, favorites, shares, publish time
- Viral pattern analysis: title feature table, content topic table, and referenceable highlights (reference notes include jump links)
- Category query prompt: lists 25 selectable categories
- Function selection entry: subscribe to push notifications / export HTML file package
Use Cases
Scenario 1: Dark horse topic mining
Role: Xiaohongshu creator
Need: Wants to find content directions that are currently gaining traction, and keep up with traffic trends
How to use: Enter your content category (e.g. "makeup and beauty") → browse the TOP50 ranking → refer to the title templates and content features in the viral pattern analysis to quickly follow up.
Scenario 2: Competitor dark horse monitoring
Role: Content operations / brand
Need: Understand which accounts in a certain track are suddenly gaining traction, and judge whether they are worth collaborating with
How to use: Select a target category and query regularly → pay attention to accounts and their content topics that appear repeatedly in the ranking → compare different statistics windows to identify dark horse accounts with sustained explosive growth.
Scenario 3: Viral pattern research
Role: Content strategy analyst
Need: Systematically study the dark horse viral post patterns of a certain category
How to use: Adjust the statistics window start date and compare rankings from different periods → analyze changes in title type proportions and content topics → export file packages for archiving, making it easier to compile analysis reports.
Notes
- Data authenticity: All values come from the data service return and are presented as-is; estimation or fabrication is strictly prohibited
- No replacement when a query yields no results: When the result is empty, state so truthfully and provide possible adjustment directions (change category or use a more recent window); do not automatically query other categories
- Credit protection: Each successful query may consume credits; unless explicitly specified by the user, do not proactively batch-query multiple categories
- Authentication: First-time use requires configuring the environment variable
REDFOX_API_KEY. For obtaining and configuring the API Key, seereferences/core_workflow.md - Link integrity: Note and author links in the output remain complete and clickable; do not use URLs truncated by the terminal