Introduction

In A-share short-term trading, developers often encounter a problem: the more detailed the screening criteria become, the more the selected stocks resemble “hindsight explanations” (look-ahead bias), often falling back immediately when chased in live trading. A common practice is to pile up volume and price conditions for a single filter without a specific sequence or a quantitative judgment on whether “the pullback is complete,” resulting in unstable quality for the candidate list.

The wrtinsd21 introduced below adopts a two-stage process: first, it uses Tonghuashun iWenCai to perform a preliminary screening based on 11 criteria to narrow down the candidate scope; then, it runs a chip distribution calculation for each candidate to determine the 90% chip concentration and convergence trend, finally providing a short-term cost-performance ranking. The responsibilities of the two steps are distinct: the first step is responsible for “finding candidates,” and the second step is responsible for “judging quality.”

What is this

wrtinsd21 is a community Skill designed for A-share short-term screening scenarios. Its coordinates are @user_201a602c/etyun54w1, published by user_201a602c, sourced from the community, categorized as professional, and is version 1.0.2.

It solves the problem of integrating the pattern of “volume increase pull-up followed by volume decrease pullback” with the quality indicator of “chip concentration” into a single screening process, outputting a ranked candidate list rather than just a filtered result.

According to SkillHub page data, this Skill has a rating of 100,000 and 1,189 downloads. Its tags include requires_api_key=true, meaning an external API Key is required at runtime.

Core Functions

This Skill executes in two steps, and both are indispensable.

Step 1: Preliminary Screening. It calls Tonghuashun iWenCai, relying on the hithink-astock-selector skill, to perform a preliminary screening of the entire A-share market based on 11 criteria, obtaining a set of candidate targets. This step does not judge the quality of the pullback; it only serves to narrow the scope.

Step 2: Chip Distribution Calculation. For each candidate from the first step, it uses a chip distribution algorithm based on a volume-price decay model to calculate the 90% chip concentration and chip convergence trend, and provides a short-term cost-performance ranking accordingly. This step strictly prohibits the use of the iWenCai API and does not depend on the data source from the first step.

The output result includes the 90% chip concentration, convergence trend judgment, and the ranked short-term cost-performance ordering for each candidate.

Installation and Activation

The captured materials do not provide the official installation command, directory structure, or specific configuration file fields for this Skill, so no unverified commands are listed here. For installation, please refer to the actual instructions on the official address page.

Official Address: https://api.skillhub.cn/user_201a602c/etyun54w1
SkillHub Page: https://www.skillhub.cn/skills/user_201a602c/etyun54w1

Typical Usage

Before using this Skill, you need to prepare a Tonghuashun iWenCai API Key. The Key must be passed via environment variables or command-line arguments at runtime; it cannot be hardcoded into the Skill.

Pass via environment variable:

export IWENCAI_API_KEY="your_key"

Pass via command-line argument:

--api-key "your_key"

Choose one of the two methods; they have the same effect and only apply to the current run. The Skill itself does not save or embed the Key.

regarding the execution flow, first perform the preliminary screening (Step 1) to get the candidate set; then execute the chip distribution calculation and ranking (Step 2) for each candidate in the set. Intermediate results should be passed between the two steps to avoid duplicate requests to iWenCai.

Applicable Scenarios and Notes

It is suitable for developers or quantitative script writers who need to screen for the “volume increase pull-up followed by volume decrease pullback” pattern in A-shares and wish to simultaneously rank quality using chip concentration. Its value lies in separating pattern conditions and chip conditions into two phases, facilitating separate parameter tuning and replacement.

Limitations to note:

  1. This Skill depends on Tonghuashun iWenCai; the preliminary screening in Step 1 must run in an environment where iWenCai is accessible.
  2. The calculation in Step 2 explicitly disables the iWenCai API. If you only want to verify the chip algorithm part, you need to bypass the first step and provide candidate targets yourself.
  3. The specific composition of the 11 preliminary screening criteria is not given in the materials; usage should be based on the actual Skill content.
  4. Rating, download counts, etc., are data displayed on the SkillHub page and will change over time; refer to the page values at the time of citation.
  5. requires_api_key=true; Step 1 cannot be executed without a configured Key.

Conclusion

wrtinsd21 is positioned to split A-share short-term screening into two segments: “preliminary screening + quality judgment,” allowing pattern conditions and chip conditions to play their respective roles, outputting a directly sortable candidate list. It does not replace market data systems but is responsible for adding a layer of screening and ranking within existing data flows.

Official Address: https://api.skillhub.cn/user_201a602c/etyun54w1