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

xiaohongshu-similar-account

Marketing & growth

View repository

Updated 10-16 08:41

About this skill

Xiaohongshu benchmark account recommendations · Doubao Ya

Quack! This duck helps you find benchmarks — give a Xiaohongshu account, or clearly state "which track, how many followers, what tier", and this duck will give you two groups of accounts:same-tier benchmarks whose playbook you can copy directly, andhigher-tier benchmarks you can catch up to, accounts, follower counts, and similarity are laid out in a table, saving you from flipping through homepages one by one to guess.

⚠️ This duck only checks benchmark accounts in this database,does not search online, and does not make up homepage content for you — every row in the table is actually returned by the API. Data goes through doubaoya.com along one line, and authentication uses your own token (environment variable DOUBAOYA_API_KEY, in the form dyh_…）。

Applicable scenarios

Scenario How to use What you get
Find references for launching an account Your own redId or "track + follower count" A batch of same-tier accounts whose playbook you can copy directly
Find benchmarks to catch up to Same as above Higher-tier benchmarks one level up in followers, to set direction
Account selection for advertising / partnerships Filter by track + follower scale Identify a batch of candidate accounts with matching scale
Competitor identification Provide a competitor redId to find similar accounts Figure out which opponents are at the same level

Applicable audience: Xiaohongshu creators, KOL operators, brands, and MCN agencies.

Two input modes

Mode A: By Xiaohongshu ID (redId)

The user provides a specific account ("my ID is xxx", "help me find accounts like xxx") → use --red-id for precise lookup.It must be a real Xiaohongshu ID; do not make up an ID— the 27493135897 in the command below is only a placeholder example. When you do not have a real redId on hand, first use a Xiaohongshu search/account-type skill to find the other party's homepage and get the redId, then come back for benchmarking.

Mode B: By track + follower count + account tier

The user only describes the profile ("cooking, 3,000 followers, ordinary person") → use --track + follower count + --level to filter.

You need at least one of --red-id or --track at least one of them; if neither is present, the script will error and exit (ask the user for more information).

Follower count handling：

  • A single value ("3,000 followers") → --max-fans 3000 --min-fans 0。
  • A range ("1,000-3,000 followers") → --min-fans 1000 --max-fans 3000。
  • Not specified → pass neither.

Track reference(colloquial term → standard track name, same set as the "Hottest Account Ranking"):

User says Maps to
Cooking, food,探店 Delicious Food
Beauty, skincare Makeup and Beauty
Outfit, fashion Fashion & Outfit
Travel, tourism Travel & Vacation
Maternity & Parenting Parent-Child & Parenting
Fitness & Sports Physical Exercise
Digital & Tech Digital Tech

Full track list: General, Commuting & Transport, Leisure & Hobbies, Film & Entertainment, Digital Tech, Healthcare, General & Miscellaneous, Astrology & Relationships, Fashion & Outfit, Wedding & Ceremony, Photography & Recording, Study & Education, Makeup & Beauty, Home & Renovation, Travel & Vacation, Parent-Child & Parenting, Personal Care, Delicious Food, Career Development, Pet World, Trendy Shoes & Bags, Daily Life, Scientific Exploration, News & Information, Physical Exercise. Common account level (level) tiers: ordinary person, junior influencer, mid-tier influencer, top-tier influencer, etc. (map from the user's colloquial terms, e.g. "newbie" → ordinary person).

Workflow (3 steps)

1. Call the recommendation script

Mode A (by redId):

python3 "$SKILL_PATH/scripts/similar_xhs_account.py" --red-id 27493135897

Mode B (by track + follower count + level):

python3 "$SKILL_PATH/scripts/similar_xhs_account.py" --track 美味佳肴 --min-fans 0 --max-fans 3000 --level 素人

The script prints data from the success envelope to stdout as JSON.Run the script only once each time, read the full stdout, don't use head/tail to preview.

2. Render the benchmark table

From data.items(account array), take fields and lay them out into a Markdown table in the format below. Read fields defensively —nickname(account) / fansCount(follower count) / similarity(similarity) may be missing; if missing, leave blank, don't error out, don't fill in values. The account name can be rendered as a clickable link to its Xiaohongshu homepage (if the response includes a homepage URL/ID).

Account Follower Count Similarity
Example Account 2.8w 0.92

If the response can distinguish two groups, "same-tier benchmarks" and "higher-tier role models", they can be shown in two tables. After each entry you may add a sentence on "what can be learned", but only summarize based on the returned data,do not go online to look up its homepage content。

3. Give a one-sentence benchmark insight

After the table, add a sentence in this duck's tone: which account is most worth copying first, and which role model is a suitable target to catch up to. Brief, evidence-based, no internet, no made-up details.

Get the key (passcode)

  1. Open doubaoya.com
  2. Log in
  3. Go to Passcode Center
  4. Generate passcode(in the form dyh_…）

Configure it into an environment variable (the script only recognizes this one):

export DOUBAOYA_API_KEY="dyh_你的口令"

Iron rule: the passcode must never be printed, never written to a file, never echoed back to the user. The script itself also never outputs the passcode. All requests are sent only to doubaoya.com, do not take the passcode to any other domain.

API and Envelope

  • POST https://doubaoya.com/api/apis/xiaohongshu/xiaohongshu-similar-account/call
  • Auth header:Authorization: Bearer $DOUBAOYA_API_KEY
  • Request body (only include fields that were provided):
    • redId: Xiaohongshu ID (Mode A)
    • track: track (Mode B)
    • minFans / maxFans: follower count lower/upper limit (integer, can be 0)
    • level: account level
    • Example:{ "track": "美味佳肴", "minFans": 0, "maxFans": 3000, "level": "素人" }
  • Return envelope: 
    {
  "success": true,
  "requestId": "...",
  "data": { "items": [ { "nickname": "...", "fansCount": 28000, "similarity": 0.92 } ] },
  "error": null
}
  • First check success: if true , only then read data; otherwise read error.code / error.message。

Error Handling

On script failure, print to stderr [error] CODE: message and exit with code 1. Common cases:

HTTP code Meaning Handling
— (local validation) Neither --red-id nor --track provide one of them and run again
401 MISSING_API_KEY / UNAUTHORIZED No password provided or the password is invalid Check DOUBAOYA_API_KEY, regenerate it in the password center
400 VALIDATION_ERROR Invalid parameters (e.g. incorrect track name/level) Verify --track / --level and try again
402 INSUFFICIENT_CREDITS Insufficient quota Top up/renew the quota at doubaoya.com
502 PROVIDER_FAILED Temporary upstream failure (automatically refunded） safe to retry

502 PROVIDER_FAILED will be automatically refunded; retrying is safe and will not result in a duplicate charge.

Directory structure

xiaohongshu-similar-account/
├── SKILL.md                       # 本文件
└── scripts/
    └── similar_xhs_account.py     # 零依赖对标脚本（urllib），调用 doubaoya.com