Query the TikTok works list by topic, presenting core data and clickable work links in descending order by play count by default.
Trigger Scenarios
Use this Skill when the user raises the following requests:
- "Query the works data for TikTok topic ID"
- "Check the works under the #xxx topic"
- "Query the topic data in a certain TikTok video"
- "Look at the works with the highest play count under a certain topic"
If the user only provides a TikTok work link, you should first extract the topic ID from the work page; if the work has no topic, directly state "no topic ID available" and do not switch to querying other topics.
Data Description
Output Content
- Works list: Output the works under the topic in table form, including sequence number, author, work ID, plays, likes, comments, shares, work link, content summary, topic
- Work link: Use a full clickable link to go directly to the work
- Pagination information: Include the number of items on this page, whether there is a next page, and the next page cursor, so you can continue viewing as needed
- Fixed reminder: Include an API Key acquisition reminder in the output
Sorting and Criteria
- By default, output in descending order by work play count; when the user specifies sorting, prioritize suggesting sorting by work metrics: plays, likes, comments, shares
- Author-level fields (follower count, following count, work count, etc.) are often measured as 0 in this data source, are not displayed by default, and are not used as a basis for judging account scale
One-Click Installation
Prerequisites
- Python 3.8+ (only uses 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 it directly
Workflow
The core execution flow (intent recognition, script invocation, table rendering rules, empty results and error handling, etc.) is detailed in
references/core_workflow.md. The Agent must follow all rules in that file when executing.
Usage Guide
Describe query requirements directly in natural language; no need to memorize commands.
| Intent | Example Phrasing | Effect |
|---|---|---|
| Query topic works | "Check the works for topic 2525872, topic name charlidamelio" | Output the works list for that topic (descending by play count by default) |
| Specify sorting | "Sort by like count and take a look" | Switch to output in descending order by like count |
| Continue paging | "Look at the next page" | Continue outputting the next page based on the previous page cursor |
| Query topic from work link | "Check the topic data for this video" | First extract the topic ID from the work page, then query |
| View raw data | "Output the raw response" | Output the raw data content |
Notes
- Data authenticity: All values come from the data service response and are displayed as-is; estimation or fabrication is strictly prohibited
- No substitution for failed queries: When there are empty results or no topic ID is available, state this truthfully and provide optional adjustment directions; do not automatically switch to querying other topics
- Credit protection: Each successful query may consume credits; unless explicitly specified by the user, do not proactively paginate or batch query multiple topics
- Authentication: First-time use requires configuring the environment variable
REDFOX_API_KEY; for API Key acquisition and configuration, seereferences/core_workflow.md - Link integrity: Work links in the output remain complete and clickable; do not use URLs truncated by the terminal