SuperAPI AI Directory
Partner中文AI newsFavorites
Skillswechat-similar-account

wechat-similar-account

General & other

View repository

Updated 10-16 04:56

About this skill

WeChat Official Account Similar Account Recommendations

Introduction

WeChat Official Account Similar Account Recommendations is an intelligent benchmarking and matching tool designed specifically for WeChat Official Account creators, helping users quickly find benchmark accounts they can reference and top benchmarks they can catch up to.

Through simple name or ID input, you can:

  • 📊 Query basic account information + data from the latest 5 articles
  • 🎯 Match peer benchmark accounts (whose strategies can be directly copied)
  • 🚀 Recommend advanced benchmark accounts (with mature models worth catching up to)
  • 📈 Generate personalized recommendation reasons based on a 3-layer weighted system

Suitable for scenarios such as WeChat Official Account creators, content operations, and account placement decisions that require benchmarking references. Powered by RedFox data platform.

Features

🎯 Core Features

  • 🔍 Intelligent Query: Supports querying by WeChat Official Account name, WeChat Official Account ID, and account category in three ways
  • 📊 Account Diagnosis: Displays the basic information, RedFox Index, reading data, and the latest 5 articles of the queried account
  • 🎯 Peer Benchmarking: Recommends accounts in the same track with the closest read counts, whose operational strategies can be directly copied
  • 🚀 Advanced Benchmarks: Recommends mature accounts with 3-5 times the read counts, whose models can be referenced and caught up to

✨ Highlights

  • ⚖️ 3-Layer Weighted Matching: Core foundation (40%) + operations and monetization (35%) + data characteristics (25%), delivering scientific and actionable recommendation results
  • 💡 7-Dimension Recommendation Reasons: Covers viral article title citations, posting time patterns, sharing and dissemination power, interaction rate, etc., and automatically supplements dimensions such as the RedFox Index and interactions over the past 7 days when data is sparse
  • 🔄 Subscription Push: Supports subscribing to benchmark account pushes, with the latest data updated daily
  • 📱 Multiple Input Methods: Flexible queries by name/ID/category, with combined queries being more precise

One-Click Installation

Prerequisites

  • Python 3.x runtime environment (only relies on the standard library, no pip install required)
  • RedFox Hub API Key (go to RedFox official website to register and obtain it; new users receive free credits)

Get API Key

  1. Visit RedFox Hub official website to learn about service details
  2. Go to registration page to register an account
  3. Newly registered users will receive free credits, and can immediately start using the API service
  4. After registering and logging in, obtain the API Key in the personal center, in the format ak_xxxxxxxx

Configure API Key

REDFOX_API_KEY obtained from environment variables. If not set, the Agent should proactively help the user configure it:

System Configuration Method Verification Command
Windows [Environment]::SetEnvironmentVariable("REDFOX_API_KEY", "<值>", "User") echo %REDFOX_API_KEY%
macOS/Linux echo 'export REDFOX_API_KEY=<值>' >> ~/.zshrc && source ~/.zshrc echo $REDFOX_API_KEY

After configuration, the terminal needs to be restarted for it to take effect, ensuring that it can also be read when switching to another skill.

Environment Variable Configuration

Variable Name Required Description
REDFOX_API_KEY is the Honghu Hub API access key, format ak_xxxxxxxx

Usage Guide

Basic Usage

1. Query by name

Tell the assistant the name of the official account you want to query:

User: Help me find benchmark accounts for "Tech Frontier"

Assistant: Call the script python scripts/wechat_similar_accountr.py --account_name "科技前沿", output the result according to the template

2. Query by ID

Provide the official account ID for a precise query:

User: Help me find benchmarks for the official account with ID gh_xxx

Assistant: Call the script python scripts/wechat_similar_accountr.py --account_id "gh_xxx", output the result according to the template

3. Query by category

Specify the account category to find benchmarks:

User: Recommend benchmark accounts in the tech and digital category

Assistant: Call the script python scripts/wechat_similar_accountr.py --account_type "科技数码", output the result according to the template

Advanced Usage

4. Combined query

Specify both name and category at the same time for a more precise query:

python scripts/wechat_similar_accountr.py --account_name "科技前沿" --account_type "科技数码"

5. Notes on querying by category

The category system of the Honghu Data platform differs from natural language,--account_type the query success rate is relatively low. When a user queries with a natural language category (such as "landscape" or "emotion"), you should:

  1. Find representative official accounts in that field (such as "landscape" → "China National Geographic", "emotion" → "Yeting")
  2. Use --account_name "代表性名称" to execute the query
  3. Confirm the platform's internal classification from the "Account Category" field in the returned results

Command Quick Reference

Command Function
--account_name "名称" Query by official account name
--account_id "ID" Query by official account ID
--account_type "分类" Query by account category

Output Flow

For the detailed core workflow (input parsing, script invocation, output templates, no-data handling, benchmark matching rules, etc.), please refer to references/core_workflow.md。

Output Order：

  1. Query basic account information + the latest 5 articles
  2. Opening remarks (only show groups with data)
  3. Same-tier benchmark table (only displayed if there is data)
  4. High-tier benchmark table (only displayed if there is data)
  5. Analysis summary (only displayed if there is data)
  6. Subscription service (must be output, regardless of whether there is data)
  7. Enterprise procurement guidance (must be output, regardless of whether there is data)

Use Cases

Scenario 1: Reference for launching a new account

Role: novice official account operator

Need: just created an official account and does not know how to position it or choose topics

How to use：

  1. Enter your own official account name or category
  2. Check the content direction and update rhythm of same-tier benchmark accounts
  3. Copy the operating strategies of benchmark accounts

Expected Benefits: quickly find replicable operating models and reduce the trial-and-error cost of launching an account

Scenario 2: Content topic optimization

Role: official account content operator

Need: content has hit a bottleneck and needs reference to the topic directions of excellent accounts in the same track

How to use：

  1. Query benchmark recommendations for your own account
  2. Analyze the viral article characteristics and content style of high-tier benchmarks
  3. Adjust your own topic strategy

Expected Benefits: Break through content bottlenecks, improve article readership and engagement metrics

Scenario 3: Account Placement Decision

Role: Brand Marketing Manager

Need: Select suitable Official Accounts for advertising placement

How to Use：

  1. Query Official Accounts in target tracks by category
  2. Compare the data performance of same-tier and higher-tier accounts
  3. Evaluate engagement rate and user persona match

Expected Benefits: Precisely select accounts for placement, improve placement ROI

Scenario 4: Competitor Analysis

Role: MCN Operations Staff

Need: Understand the operational strategies and data performance of competitor accounts

How to Use：

  1. Query benchmark recommendations for competitor accounts
  2. Analyze the differentiated positioning of accounts in the same track
  3. Discover market gap opportunities

Expected Benefits: Grasp competitor dynamics, discover differentiated competitive opportunities

Project Architecture

Directory Structure

wechat-similar-account/
├── scripts/
│   └── wechat_similar_accountr.py   # 核心脚本：调用API查询对标账号并格式化输出
├── references/
│   └── core_workflow.md              # 核心工作流程：操作步骤、输出模板、匹配规则
├── SKILL.md                          # 技能描述文件
└── README.md                         # 项目说明文档

Tech Stack

Component Description
Runtime Environment Python 3.x
Dependencies Python standard library (json, argparse, os, urllib, platform, re)
Data Source Red Fox Data API (https://redfox.hk?source=github）
API Interface POST https://redfox.hk/story/api/gzhUser/querySimilarAccounts

Core Module Description

Module Responsibility
query_similar_accounts() Call the Red Fox Data API to query benchmark accounts and top accounts (API returns currentAccount + benchmarkAccounts + topAccounts)
format_output() Format the complete text output, including queried account information, benchmark table, and analysis summary
generate_recommendation_reason() Generate recommendation reasons based on a 3-layer weighted matching system, supporting combined output of 7 dimensions
_analyze_publish_schedule() Infer fixed posting time slots from article publishTime (e.g., "fixed posting at 7 AM" "mainly evening time slot")
_extract_top_article() Extract the title of the most-read article, output "hit article 「title…」 reaches X.X times the average readership"
_calc_interaction_rate() Calculate engagement rate; if it exceeds 100%, determine it as abnormal data and do not output
_calc_share_rate() Calculate share rate = shares / reads, reflecting content spreadability
calc_seven_day_reads() Calculate reads in the last 7 days (accumulate clicksCount from works; return 0 if works is empty; interactiveCountSeven is engagement volume, not reads, and must not be mixed up)

API Response Structure

API (POST /gzhUser/querySimilarAccounts) Return the following fields：

  • currentAccount: Basic information of the queried account and data of the last 5 articles (works field)
  • benchmarkAccounts: List of same-tier benchmark accounts
  • topAccounts: List of higher-tier benchmark accounts

The script directly retrieves currentAccount to obtain the queried account information and works data. When currentAccount is empty (e.g., when querying by category does not hit a specific account), match by accountName/accountId from benchmarkAccounts + topAccounts.

Key Field Description

Field Description
articleCountSeven Number of articles published in the last 7 days; currentAccount may not contain this field, in which case infer from the number of works
interactiveCountSeven Last 7 daysengagement volume(not reads), must not be used as reads
avgReadCount Average reads, may be null, in which case effective_avg needs to be calculated from works
clicksCount Article read count, maximum value is 100001 (100k+ capped value)
redfoxIndex RedFox Index, overall account quality score
works Recent article list, including title/clicksCount/likeCount/commentCount/watchCount/interactiveCount/shareCount/publishTime/workUrl, etc.

Recommendation reason generation dimensions

Priority Dimension Data source Example
1 Viral article insights + content topic focus works + effective_avg 3 viral articles in the same track in the past 7 days, all focused onParenting and child education/Health and wellness
2 Viral article title citation The most-read article in works Viral article "The best menopause period for women..." reached 2.4 times the average read count
3 Update cadence + content strategy + posting time articleCountSeven + works High-frequency daily updates, in-depth/medium-depth graphic and text content, fixed morning posting at 7 AM
4 Interaction rate + share rate Interaction/share/read data in works Interaction rate 3.2%, share rate 5.1%, strong content dissemination power
5 Interaction count in the past 7 days interactiveCountSeven 498k interactions in the past 7 days, user activity can be used as a reference
6 RedFox Index stage positioning redfoxIndex + accountType RedFox Index 932, overall account quality in "Digest selections" track performs outstandingly
7 Sparse data supplementation Multi-dimensional fallback Content direction: parenting/health and wellness, 5 articles published in the past 7 days

Key mechanisms：

  • effective_avg = max(avgReadCount, works average read count), can still determine viral articles when avgReadCount is null
  • Interaction rate exceeding 100% is judged as abnormal data and is not output
  • Automatic deduplication across dimensions to avoid repeated output of the same piece of information

FAQ

Installation-related questions

Q1: What should I do if it prompts "REDFOX_API_KEY configuration not found"?

A: Please configure it according to the following steps:

  1. Visit https://redfox.hk?source=github to register an account and obtain an API Key
  2. Windows users, execute in PowerShell:[Environment]::SetEnvironmentVariable("REDFOX_API_KEY", "<你的API Key>", "User")
  3. macOS/Linux users, execute:echo 'export REDFOX_API_KEY=<你的API Key>' >> ~/.zshrc then source ~/.zshrc
  4. restart the terminal for it to take effect

Q2: Do I need to install additional Python dependencies?

A: No. The script only uses the Python standard library (json, argparse, os, urllib, re, etc.), and no pip install of any package is required.

Usage-related questions

Q3: What should I do if I cannot query account data?

A: Please confirm whether the account name or ID was entered correctly. If the account is new or has published few articles recently, complete data may temporarily be unavailable; it is recommended to try again later.

Q4: What query methods are supported?

A: Three methods are supported:

  • By official account name:--account_name "名称"
  • By official account ID:--account_id "gh_xxx"
  • By account category:--account_type "科技数码"
  • Combined queries are also supported, such as specifying both name and category

Q5: What is the difference between same-tier benchmarks and higher-tier benchmarks?

A: The read counts of same-tier benchmarks are closest to the queried account, so their methods can be directly copied; the read counts of higher-tier benchmarks are 3-5 times those of the queried account, and their operation models are more mature, making them suitable for catch-up learning.

Q6: What should I do if a category query returns no results?

A: There are differences between the Red Fox Data platform's classification system and natural language. It is recommended to search using the names of representative official accounts in that field. For example, "Scenery" → "Chinese National Geography", "Emotion" → "Night Listening".

Troubleshooting

Q7: What should I do if an API request returns an HTTP error?

A: Please check the following points:

  1. Confirm that the API Key is correct and has not expired
  2. Confirm that the network can access https://redfox.hk/?source=github
  3. Check whether the API Key has a sufficient credit balance

Q8: Are some fields in the output result empty or 0?

A: This is normal and indicates that the account currently has no data records for the corresponding data dimension. The script will automatically handle it as "No data" or "0" for display.

Get Help

If you have other questions, you can get help through the following methods: