SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillslinkfox-chuhaijiang-tiktok-shop

linkfox-chuhaijiang-tiktok-shop

Data & analytics

View repository

Updated 10-16 00:00

About this skill

TikTok Shop Store Intelligence

Use this skill to research public TikTok Shop stores: discover and filter stores, inspect one store, trace its creators/products/videos, or browse most-promoted and top-selling store rankings.

Core Concepts

  • Normal endpoints require one lowercase marketplace code: br,de,es,fr,gb,id,it,jp,mx,my,ph,sg,th,us,vn.
  • Search and relationship endpoints return at most 10 records per page; ranking endpoints accept 1–20.
  • Store detail may return data.items, data.core.items, and data.channel.items. When include=core,channel, read all three branches.
  • Monetary fields are usually { "unit": "US", "value": 31 }. Preserve the returned unit and inspect the runtime type before formatting.
  • Each endpoint call is billed independently. Call only the endpoint needed for the current question; successful calls use a 24-hour local cache keyed by endpoint, gateway, API Key fingerprint, and parameters.

Data Fields

Field Meaning
data.total_count Total matching records
data.items Store, creator, product, video, or ranking rows
data.core.items Optional store scale and 30-day core metrics
data.channel.items Optional 30-day video/live/product-card/creator channel metrics
id / seller_id Store or related-entity identifier
shop_name / region Store identity and marketplace
shop_rating Store rating
shop_product_count Current product count when supplied
shop_total_sold_count / shop_total_gmv Lifetime sales and GMV when supplied
shop_sold_count_for_last_7_days / shop_gmv_for_last_7_days Recent store performance
interval_* Metrics for the requested ranking window

Relationship rows use endpoint-specific tiktok_seller_* fields. Read references/api.md for the exact request definitions, production-validated response paths, and observed field catalog. Do not assume one shared row schema.

Endpoints

Intent Endpoint Script Points/call
Filter and discover stores /chuhaijiang/sellers/search chuhaijiang_seller_search.py 18
Inspect one store /chuhaijiang/sellers/detail chuhaijiang_seller_detail.py 9
Find associated creators /chuhaijiang/sellers/related-creators chuhaijiang_seller_related_creators.py 18
List store products /chuhaijiang/sellers/related-products chuhaijiang_seller_related_products.py 18
Find associated videos /chuhaijiang/sellers/related-videos chuhaijiang_seller_related_videos.py 18
Browse most-promoted stores /chuhaijiang/sellers/rankings/most-promoted chuhaijiang_seller_rank_most_promoted.py 18
Browse top-selling stores /chuhaijiang/sellers/rankings/top-selling chuhaijiang_seller_rank_top_selling.py 18

Route one intent to one endpoint. Use a store id from search or ranking results for detail and relationship calls. Ask before making additional paid drill-down calls the user did not request.

Parameter Guide

  • country: required lowercase marketplace code.
  • id: required for detail and relationship endpoints; send it as a string to preserve precision.
  • page: starts at 1. pageSize is optional and must not exceed the endpoint limit.
  • Search accepts keyword, category, sellerType, rating/7-day sales/7-day GMV ranges, and sort (field:asc or field:desc).
  • Detail accepts include=core, include=channel, or both values in either comma-separated order (core,channel / channel,core).
  • Both rankings require date (YYYYMMDD) and granularity (daily, weekly, monthly, 0, 1, or 2); optional shopType is one of 1,2,3,4.
  • Use the camelCase request names in references/api.md; do not send the upstream snake_case aliases.

How to Call

  • API Endpoints：POST /chuhaijiang/sellers/search、POST /chuhaijiang/sellers/detail、POST /chuhaijiang/sellers/related-creators、POST /chuhaijiang/sellers/related-products、POST /chuhaijiang/sellers/related-videos、POST /chuhaijiang/sellers/rankings/most-promoted、POST /chuhaijiang/sellers/rankings/top-selling(for full parameters/responses/error codes, see references/api.md）
  • Python Script: use respectively python scripts/chuhaijiang_seller_search.py '<JSON 参数>' [--inline]、chuhaijiang_seller_detail.py、chuhaijiang_seller_related_creators.py、chuhaijiang_seller_related_products.py、chuhaijiang_seller_related_videos.py、chuhaijiang_seller_rank_most_promoted.py、chuhaijiang_seller_rank_top_selling.py
  • Cost Constraints: store details cost 9 compute units per call; the other six endpoints cost 18 compute units per call. By default, the business entry script caches only successful responses for 24 hours; business-successful empty results can also be cached; the cache is isolated by the current working directory, calling identity, gateway, endpoint, and complete request parameters. When the same combination hits, it outputs Cache hit, and no new paid request will be made. HTTP or business failures are not cached;--inline does not bypass the cache;--no-cache forces a real request and may consume compute units again, so use it only when the user explicitly agrees to additional consumption or when debugging.

Output Strategy (Default Script Behavior)：

  • Alwayswrite the complete response to <cwd>/linkfox/<YYYY-MM-DD>/<session>/data/linkfox-chuhaijiang-tiktok-shop-<timestamp>.json（<cwd> is the working directory when the script is executed, which in Claude Code is the current project directory;<session> is taken from the environment variable SESSION_ID, automatically aggregated by user task;writing to /tmp is prohibited, and an error is reported if the current directory is not writable)
  • Response body ≤ 8 KB: after writing to disk, print the complete JSON to stdout
  • Response body > 8 KB: after writing to disk, stdout outputs only a summary (top-level fields, business total count, length of the largest list field + first 3 samples)
  • Add --inline to force full printing to stdout (also written to disk)

Data Reading Suggestions: first look at the summary to judge whether it is sufficient; when specific fields are needed, prefer using jqorConvertFrom-Json to extract on demand from the saved json file, avoiding bringing the entire JSON into context.

On Windows hosts whose Python stdout still uses GBK, set PYTHONIOENCODING=utf-8 before printing rows that may contain emoji. This affects console rendering only; the UTF-8 JSON file is already saved first.

Resolving Authentication and Compute Unit Issues

When the following abnormal situations occur, use references/onboarding.md to guide resolution:

Abnormal Situation

  • API Key not configured: the environment variable is not configured LINKFOX_AGENT_API_KEY, and LINKFOXAGENT_API_KEY。
  • Response 401 or 402 status code
  • Response indicates insufficient compute units or balance: the message contains "insufficient compute unit balance/insufficient billing/insufficient balance/quota exceeded/insufficient balance/package expired/recharge required/please recharge", or similar meaning.

Usage Examples

python scripts/chuhaijiang_seller_search.py '{"country":"us","keyword":"beauty","minRating":4,"page":1,"pageSize":3,"sort":"gmv_7d:desc"}'
python scripts/chuhaijiang_seller_detail.py '{"country":"us","id":"7495205878591949358","include":"core,channel"}'
python scripts/chuhaijiang_seller_related_creators.py '{"country":"us","id":"7495205878591949358","page":1,"pageSize":10}'
python scripts/chuhaijiang_seller_related_products.py '{"country":"us","id":"7495205878591949358","page":1,"pageSize":10}'
python scripts/chuhaijiang_seller_related_videos.py '{"country":"us","id":"7495205878591949358","page":1,"pageSize":10}'
python scripts/chuhaijiang_seller_rank_most_promoted.py '{"country":"us","date":"20260824","granularity":"daily","pageSize":10}'
python scripts/chuhaijiang_seller_rank_top_selling.py '{"country":"us","date":"20260824","granularity":"daily","pageSize":10}'

Display Rules

  1. State the marketplace, filters, current page, total count, and ranking date/granularity used.
  2. For store lists, show store name, rating, product count, recent or interval sales/GMV, creator reach, region, and store ID when available.
  3. For detail, group base profile, core metrics, and channel metrics; do not ignore data.core.items or data.channel.items.
  4. For creators, products, and videos, show the related entity ID plus its returned performance metrics; retain the store ID for context.
  5. Preserve currency units and field time windows. Do not relabel unit: US or mix lifetime, 7-day, 30-day, and ranking-window values.
  6. Never expose API keys. Treat URLs as expiring when the response contains signed query parameters.

Important Limitations

  1. This is public market intelligence, not private seller-center data.
  2. It does not manage orders, inventory, listings, fulfillment, affiliates, advertising, or store authorization.
  3. Some fields may be absent or null; preserve unknown fields and inspect runtime types before rendering.
  4. Search and rankings change over time. Report the request date and ranking window with results.
  5. sellerType belongs to search, while shopType belongs to rankings. Do not interchange them.
  6. Do not invent unsupported filters or compensate by repeatedly calling other paid endpoints.

User Expression & Scenario Quick Reference

Applicable:

  • Store Discovery and Details: "Use Chuhaijiang to filter US TikTok beauty stores", "Use Chuhaijiang to view this store's details".
  • Relationship Drilling: "Check the creators this TikTok store collaborates with", "View in-store products", "Find this store's sales videos".
  • Store Rankings: "Use Chuhaijiang to view the US TikTok store trending ranking", "Use Chuhaijiang to check the TikTok store sales ranking".
  • Combined Analysis: "First filter stores with high 7-day GMV, then analyze their creators, products, and videos".

Adjacent: General TikTok store ranking or detail requests without a source should use linkfox-kalodata-tiktok-shop. Use the EchoTik seller skills when the user asks for EchoTik fields or multi-period seller profiles. Use official TikTok Shop skills for the user's authorized seller-center operations.

Not applicable: order handling, stock or listing changes, fulfillment, store authorization, ad campaign changes, non-TikTok marketplaces, or a request naming another data provider.

Compute Unit Consumption Rules

Billing is per endpoint: shop details 9 compute units per call; shop search, associated creators, associated products, associated videos, and the two types of rankings 18 compute units per call.

Users will be charged for compute unit consumption. Before high-frequency calls or multi-endpoint drill-downs, first explain the estimated number of calls and compute unit consumption.

Feedback: Use the Feedback API in references/api.md only when the user explicitly asks to submit feedback or explicitly authorizes submission after an issue is identified. Summarize and sanitize the content; never include API keys, phone numbers, raw user messages, or unrelated task context.

For more high-quality, professional cross-border e-commerce skills, visit LinkFox Skills.