SuperAPI AI Directory
Partner中文AI newsFavorites
Skillsx-top-account

x-top-account

Data & analytics

View repository

Updated 10-16 07:27

About this skill

X (Twitter) Popular Account Ranking

1. Introduction

X (Twitter) daily most influential account ranking tracking and analysis tool -- based on the Red Fox Data API, it provides daily influence account ranking queries for the X platform across all industries / by industry / by gender, and supports ranking queries, filtering, pagination, HTML visualization report downloads, and scheduled subscription push.

Target users: overseas content creators, brands / overseas advertising personnel, MCN agencies / cross-border operations teams, industry analysts / consulting practitioners.

2. Features

Core features：

  • Ranking query -- check the daily TOP account ranking (20 entries per page, 10 pages total, 200 entries)
  • Filtering -- precisely filter by gender and 32 industry categories, with natural language fuzzy matching support
  • Pagination -- support querying a specified page number and browsing page by page in the conversation
  • Report download -- generate an X platform-style HTML report, with one-click export to PDF / high-definition images
  • Scheduled subscription -- set up daily scheduled push, with support for subscriptions by gender / industry combinations

Highlights：

  • Updates the previous day's ranking every morning at 9:00, with up to 7 days of lookback
  • Influence score: a weighted comprehensive evaluation combining follower count, engagement, impressions, etc., out of 100
  • Industry fuzzy matching: supports natural language input (e.g. "pets" automatically maps to "cute pets and animals")
  • Multi-dimensional data: country/region, main industry, daily growth, X follower count, follower count on other platforms, and topic tags at a glance

3. One-click installation

Authentication

Get API Key

Please go to Red Fox hub to get the API KEY

Configure API Key

Option 1: Taking OpenClaw as an example, add REDFOX_API_KEY to ~/.openclaw/openclaw.json

{ "env": { "REDFOX_API_KEY": "ak_xxxx..." } }

Option 2: Terminal configuration: export REDFOX_API_KEY="ak_xxxx..."

export REDFOX_API_KEY="ak_xxxx..."

Dependency installation

No third-party dependencies need to be installed; it runs using the Python standard library.

4. Usage guide

Basic usage

Query ranking data (default page 1, 20 entries) and generate an HTML report：

python scripts/fetch_rank.py --query "用户原始问题" --html

The script will automatically parse:

  • Date: if not specified, take the latest available date (updated daily at 9:00; after 9:00 it is yesterday, before 9:00 it is the day before yesterday)
  • Gender: recognize male/female (e.g. "female blogger" → female); if not specified, all
  • Industry: fuzzy match 32 industries (e.g. "pets" → "cute pets and animals"); the API category passes the Chinese industry name; the skill does not expose English category values internally
  • Page number: recognize expressions such as "page 2"; if not specified, page 1

Pagination query：

python scripts/fetch_rank.py --query "用户原始问题" --page 2 --html

Explicitly specify filter conditions(higher priority than --query parsing results):

python scripts/fetch_rank.py --date 2026-09-20 --gender female --category "萌宠动物" --html

Important: The number of data entries displayed in the HTML report must match the number displayed in the conversation (20 per page). Use deliver_attachments to deliver the HTML file.

Advanced Usage

Update time and retroactive range (must be output to the user)：

  • The ranking is updated every morning at 9:00 updateprevious daydata
  • up tothe past 7 days

Time boundary handling (must be strictly followed)：

  • Query for the current day or future dates: reply "We are very sorry🙏, our latest data is from yesterday, and we will provide you with the yesterday hot list closest to your needs."
  • Query time earlier than the retroactive date: reply "We are very sorry🙏, currently the ranking supports retroactive lookup for up to 'the past 7 days'. I will query the time range closest to your needs for you~"
  • The target date API returns empty data: the script automatically falls back to the nearest available date and outputs [WARN] prompt, which must be relayed to the user

Subscription Service：

After the user confirms the subscription, use the automation_update tool to create an automated task:

  • Daily subscription:FREQ=DAILY;BYHOUR=9;BYMINUTE=30(pushed after data is updated at 9:00)

automation prompt example:查询X(Twitter)最新热门账号榜（行业：全部行业，性别：全部），生成报告并推送给用户

When asking about a subscription, all filter options must be listed(see the "Output Format" subscription service template):

  • Gender: all / male / female
  • Industry category: list all 32 industry names in Chinese; for the complete enumeration, see references/category_map.md

Common Command Quick Reference

User intent Typical trigger words Corresponding operation
Ranking query X ranking, Twitter daily ranking, popular Twitter accounts, influence ranking Call the script to query and format the output
Filter query Pet category/Tech category/Female bloggers...Twitter ranking Call the script and filter by industry/gender
Pagination page 2, next page, continue viewing Add the --page parameter and query again
Report download download report, export ranking, generate report Run the script with --html to generate and deliver

Output Format

Markdown table template：

📈 X (Twitter) 热门账号榜 · {category_label} · {gender_label}
数据日期：{rankDate}（第 {pageNum}/{pages} 页）
共 {total} 个账号上榜（本页展示 20 条）

💡 榜单说明：每日早上 9:00 更新前一天榜单，与实时数据存在差异
📐 影响力评分：结合粉丝数、互动量、曝光数等加权综合评估，满分 100

| 排名 | 账号名 | 账号简介 | 国家/地区 | 主行业 | 影响力评分 | 日涨幅 | X粉丝数 | 其他平台粉丝数 | 专题标签 |
|:---:|--------|--------|:---:|--------|:---:|:---:|:---:|--------|--------|
| 🥇 1 | [Elon Musk](profile_url)<br>*Tech Entrepreneur and CEO* | biography 简介（前 60 字） | 美国 | Tech & Software | 99/100 | +1.25% | 2.2亿+ | — | 账号标签 |
| 🥈 2 | [MrBeast](profile_url)<br>*Creator & Philanthropist* | biography 简介（前 60 字） | 美国 | Entertainment & Culture | 97/100 | +0.88% | 3,100w+ | YouTube 4亿+ | — |
| 🥉 3 | [iJustine](profile_url)<br>*Tech Vlogger* | biography 简介（前 60 字） | 美国 | Tech & Software | 93/100 | +0.69% | 180w+ | YouTube 7,100w+ / TikTok 1,600w+ | Women in Tech / STEM |

Field display rules：

  • Account name: Markdown link [fullName](profileUrl) + <br> + title (title, italic), the title must not be moved to another column or omitted; add 🥇🥈🥉 to ranks 1-3
  • Account bio: only biography(truncate to the first 60 characters if too long); do not put the title in this column
  • Influence score: xScore rounded to the nearest integer, format 99/100
  • X follower count: ≥100 million display x.x亿+, ≥10 thousand display x,xxxw+
  • Follower counts on other platforms: display each one as otherNetworks[].network platform name + followersFmt(e.g., YouTube 400w+the field name is network, not platform); connect multiple platforms with / , or if none —
  • Topic tag: take the topic topic topics; if none, take the public welfare tag cause; if neither, then —
  • Completeness: output all 10 table columns; do not omit any field or entire column to compress column width; the HTML report and the conversational table layout must remain consistent (the title is below the account name, and the bio column contains only biography)

After outputting the ranking, append:

总结榜单构成
{AI 对榜单数据做一些分析：头部集中度、行业分布、国家分布、涨粉亮点等}

⚡ 更多操作
• 点击下方下载 HTML 报告文件，可在浏览器中打开查看，支持一键导出 PDF / 高清图片
• 本次榜单完整共 {total} 条数据，是否需要继续查看第 {pageNum+1} 页？

📬 订阅服务
1️⃣ 是否需要订阅每日 X(Twitter) 账号最新排名，订阅后定时推送给您（每日早上 9:00 更新前一天榜单）。
2️⃣ 支持您按照以下筛选项订阅：
   • 性别：全部 / 男 / 女
   • 行业分类：全部行业、商业创业、汽车出行、设计创意、航空领域、电商零售、教育培训、工程运营、娱乐文化、环境能源、家庭育儿、时尚美妆、金融投资、美食饮品、游戏电竞、健康医疗、牧场马术、宗教信仰、科学学术、体育健身、科技软件、旅行户外、视觉艺术、健康生活、居家手工、人力职场、公益平权、法律法务、市场营销、音乐领域、萌宠动物、政治新闻

5. Use Cases

User Segment Core Pain Point How to Use
Overseas Content Creators Don't know who the top accounts in the same niche are, no benchmark to follow Query "Tech Twitter ranking" to get benchmark references
Brands / Overseas Advertisers Low efficiency in finding overseas influencers, unaware of accounts' cross-platform influence Query "Beauty female blogger ranking" to evaluate advertising candidates
MCN / Cross-border Operations Teams Difficult to obtain competitor account intelligence Query "Pet ranking page 2" to monitor the market page by page
Industry Analysts Single data dimension, lack of visualized reports Generate "Financial Investment Industry Ranking HTML Report"

Expected Benefits: Quickly grasp the landscape of top accounts on the X platform, identify high-growth accounts, and reduce the decision-making cost of overseas advertising.

6. Project Architecture

Directory Structure

x-top-account/
├── SKILL.md                          # 本文件
├── scripts/
│   ├── fetch_rank.py                 # 查询榜单数据（解析/翻页/容错回退）
│   └── generate_report.py            # 生成 X 风格 HTML 报告（导出 PDF/图片）
├── assets/
│   └── category_config.json          # 更新规则 + 性别/行业枚举与关键词映射
├── output/                           # HTML 报告输出目录（运行时自动创建）
└── references/
    ├── category_map.md               # 行业映射表与匹配规则
    └── api_docs.md                   # API 接口详细文档（技术参考）

Core Module Description

  • fetch_rank.py: Receives user query parameters, calls the Honghu API to obtain ranking data, automatically parses date, gender, industry, and page number, automatically falls back to the most recent available date when there is no data for the target date, and outputs a JSON data file; adding --html automatically calls report generation.
  • generate_report.py: Reads JSON data to generate an X platform-style (pure white background + X blue) HTML report. The report title format is X(Twitter)·{行业}·热门{性别}账号榜(when industry/gender is all, the corresponding field is omitted), saved by default to the skill directory under output/(automatically created), with filename format X账号榜_{性别}_{分类}_{日期}.html(when gender/category is all, displayed as all genders/all categories, e.g., X账号榜_全部性别_全部分类_2026-09-21.html). The column layout is consistent with the chat table: the account name column displays avatar + account name (clickable to jump to the X homepage) + title (below the account name, ellipsis if too long, full text on hover), the account bio column contains only biography (full text on hover), and follower counts on other platforms include the platform name and are clickable to jump to the corresponding homepage; built-in export PDF / high-definition image buttons, automatically opens the browser after generation.

Resource Index

File Purpose
scripts/fetch_rank.py Query ranking data
scripts/generate_report.py Generate HTML visualization report
references/category_map.md Industry mapping table and matching rules
references/api_docs.md Detailed API interface documentation (technical reference)

7. FAQ

Installation

Q: What dependencies need to be installed? A: No third-party dependencies need to be installed; it can run using the Python standard library.

Q: How to configure the API Key? A: Please go to Honghu Hub to obtain an API KEY, then configure it via the environment variable REDFOX_API_KEY .

Usage

Q: When is the ranking data updated? A: The previous day's ranking is updated every morning at 9:00. If the current time is before 9:00, the script will automatically use the data from the day before yesterday.

Q: What filters are supported? A: 3 gender options (All/Male/Female) + 32 industry categories. For the complete list, see references/category_map.md。

Q: What is the data lookback range? A: Up to the past 7 days; when there is no data for the target date, it automatically falls back to the most recent available date and provides a prompt.

Q: What should I do if industry recognition fails? A: When it cannot be recognized, fall back to all industries and list the complete industry list for the user to choose from.

Q: How many items per page? How to turn pages? A: A fixed 20 items per page, 10 pages total, 200 items; during the conversation, ask the user whether to view the next page, use --page N to turn pages.

Troubleshooting

Q: HTML report generation failed? A: Ensure the workspace has write permission and the JSON data file path is correct.

Q: Data inconsistency? A: However many items are displayed in the conversation (20 per page), the HTML report must display the same number; the two must be consistent. After turning pages, the report for the corresponding page must be regenerated.