Preface

Most people involved in data acquisition have gone through this process: writing scripts to scrape pages, adding code to bypass anti-scraping measures, and finally manually piecing fields back into tables. These three tasks are scattered across temporary code, debugging notes, and chat records. Whenever the target site changes or a new person takes over, the work has to be redone.

The “Crawler Data Acquisition Expert” aims to solve exactly this: packaging “Data Acquisition + Data Cleaning” into a solution with explicit trigger keywords. This allows an Agent to follow a fixed workflow when receiving acquisition requests, rather than writing scripts from scratch every time.

This article is based on official materials fetched on 2026-10-30. At the time of writing, the official page and interface were inaccessible (returning fetch failures). Therefore, the facts below are based entirely on the pre-fetched materials; unverified content has been omitted.

What is this

In one sentence: A Skill solution for crawler data acquisition and cleaning, designed for enterprises and individuals.

  • Name: Crawler Data Acquisition Expert
  • Coordinates: @kunlungrowth/kunlun-da-664
  • Version: 1.0.0
  • Source: enterprise
  • Category: data-analysis
  • Publisher: Kunlun Growth
  • Official URL: https://api.skillhub.cn/kunlungrowth/kunlun-da-664
  • SkillHub Page: https://www.skillhub.cn/skills/kunlungrowth/kunlun-da-664

Problem Solved: Data acquisition and cleaning have long relied on one-off scripts, lacking unified trigger conditions, workflows, and output conventions. This Skill uses “Trigger on” keywords to define activation timing, thereby standardizing the process.

Page Snapshot Data (Fetched 2026-10-30): Rating 100,000, Downloads 1,189, Installs 0. The zero install count implies there are currently no public adoption records to reference, and this article does not cite any user feedback.

Core Capabilities

The capabilities verifiable from the materials are limited to the two points explicitly stated in the introduction:

  1. Data Acquisition: Scraping required data from target sites.
  2. Data Cleaning: Cleaning scraped results to form usable datasets.

The activation method is keyword-driven. The official introduction provides four “Trigger on” keywords:

Data Acquisition / Crawler / Data Scraping / Data Cleaning

This means that when a request matches any of these four terms, the Skill is activated and executes the acquisition and cleaning workflow, rather than treating it as a standard Q&A.

Another notable field is the tag requires_api_key=false, which means no additional API Key configuration is required to call this Skill. For enterprises or individuals who do not wish to maintain keys, this is a threshold advantage.

Note: The materials do not contain implementation details regarding anti-scraping strategies, proxy pools, scheduling, incremental crawling, or JS rendering. Supported data source types are also not listed. This article makes no inferences regarding the existence of these capabilities.

Installation and Activation

It must be stated truthfully here: The fetched materials do not contain any installation commands, directory paths, or configuration file fields. Since the official page was inaccessible at the time of writing, this article does not provide specific installation instructions.

Only the entry points themselves can be verified:

https://api.skillhub.cn/kunlungrowth/kunlun-da-664

https://www.skillhub.cn/skills/kunlungrowth/kunlun-da-664

This Skill is based on the general SKILL.md format. Skills of this format are typically loaded as directories in Agent hosts that support skills. This involves placing the SKILL.md file into the host’s skills directory, where the host reads it when matching trigger words. If you are using Cursor, Codex CLI, or Claude Code, the specific directory names, loading commands, and configuration methods vary. These differences were not present in the fetched materials.

Therefore, the recommended operating sequence is: First, open the official URL to confirm the installation method, then configure it according to your host’s documentation. Do not simply copy paths from other Skills. If your host does not support SKILL.md loading at all, this Skill cannot be activated, and its keyword trigger mechanism will not function.

Typical Usage

The reproducible examples provided in the official materials consist only of trigger methods, without accompanying sample scripts or data. Below is the usage organized by “Trigger on” keywords.

Scenario 1: Matching “Data Scraping”

Help me scrape a product list page from a site, extracting the title, price, and listing time

This request contains “scrape,” matching the trigger word, so the Skill enters the acquisition workflow.

Scenario 2: Matching “Data Cleaning”

Clean the product data just scraped, removing duplicate rows and standardizing prices to numeric types

This request contains “clean,” also matching the trigger word, so the Skill enters the cleaning workflow.

Scenario 3: Matching both, running the full pipeline

Acquire this batch of data and clean it into a format ready for direct database entry

“Acquire” and “clean” are both in the trigger word list, so the Skill executes in the order of acquisition to cleaning.

Usage Constraints: Since activation relies on keyword matching, it is best to explicitly include words like “acquire,” “scrape,” “crawler,” or “clean” in the request. Simply saying “I want this data” without using trigger words may not activate the Skill. Additionally, users must independently judge whether the acquisition target is compliant and whether site robots and access frequency limits are met; the official materials do not provide instructions on this.

Applicable Scenarios & Notes

Suitable Scenarios:

  1. Teams or individuals requiring a fixed workflow for acquisition and cleaning, without wanting to rewrite scripts every time.
  2. Requirements that match acquisition or cleaning keywords, where the user wants the Agent to follow a plan rather than improvise.
  3. Situations where one does not want to apply for and configure a separate API Key for the call.

Limitations & Pitfalls:

  1. Opaque Implementation Details: The materials lack explanations for anti-scraping, proxies, concurrency, retries, or JS rendering. You need to confirm the capability boundaries before tackling complex sites.
  2. Zero Installs: There is a lack of public adoption feedback. If problems arise, there are no community cases to consult; you can only return to the official URL to submit requests.
  3. Keyword Trigger Dependency: Activation is string matching. If wording deviates from the trigger words, it may not activate. It is recommended to explicitly include “acquire/scrape/clean” in the requirement description.
  4. Compliance Responsibility: The responsibility lies with the user. The scope, frequency, and data usage of the acquisition must be evaluated independently. The Skill will not judge whether the target site allows scraping for you.

Conclusion

The value of the “Crawler Data Acquisition Expert” lies not in anti-scraping or cleaning algorithms, but in solidifying the workflow and trigger conditions for acquisition and cleaning into a reusable solution, thereby lowering the startup cost for every acquisition request. It requires no API Key and has clear trigger conditions. However, implementation details and installation methods are not elaborated in the public materials and require verification via the official URL first.

Official URL: https://api.skillhub.cn/kunlungrowth/kunlun-da-664

SkillHub Page: https://www.skillhub.cn/skills/kunlungrowth/kunlun-da-664