Introduction

When discussing mutual funds (public funds) in an Agent, the common practice is to paste screenshots of Net Asset Values (NAV) or let the model comment based on general common sense. Whether the data is up-to-date, the dimensions are complete, and whether “who it is suitable for” is separated from “whether there are obvious risks” often depends on the specific prompt used, making it difficult to reuse.

Below is an introduction to the “Public Fund Elevated Diagnosis” (公募基金升维诊断) on SkillHub: upon providing a fund code, it automatically fetches data, performs multi-dimensional evaluations, and outputs a target audience profile, while also serving to screen out products with obvious risk signals. The Skill explicitly does not predict future performance and is suitable for scenarios requiring structured conclusions without turning the conversation into a stock-picking script.

What is this

“Public Fund Elevated Diagnosis” is a community-sourced professional-tier Skill. The current version is v2.9.15, with coordinates @user_525a1b37/fundx, maintained by the publisher user_525a1b37.

The official description positions it as a “Public Fund Elevated Diagnosis System”: input a fund code to automatically fetch data, conduct multi-dimensional evaluation, and generate a target audience profile, used to exclude risky products; it does not predict future performance. The page is marked requires_api_key=false, meaning that according to SkillHub information, no additional API Key is required for use (though specific data still depends on the Skill’s built-in or runtime acquisition methods).

On SkillHub, this Skill has a rating of 100000 and 838 downloads (based on the SkillHub page and crawled data).

Core Functions

The following capabilities are derived from the official description and the SkillHub page; details of unpublished implementations are not expanded upon:

  1. Fetch Data by Code: After the user or Agent provides a mutual fund code, the Skill automatically fetches relevant data, eliminating the need to manually organize market tables within the conversation.
  2. Multi-dimensional Evaluation: Performs a structured evaluation of funds within a unified workflow, facilitating comparison between different products rather than a one-off, free-form long-form commentary.
  3. Target Audience Profile: Outputs descriptions of the suitable audience matched to the product’s characteristics, helping to distinguish between “how the product is” and “whether it suits the current reader.”
  4. Risk Exclusion Orientation: Emphasizes identifying and excluding products with obvious risks, rather than making profit promises or trend predictions.
  5. No Future Performance Prediction: The boundary is hardcoded into the product positioning; it is suitable for compliance and research assistance use cases and should not be treated as an investment advice engine.

Installation and Activation

Method 1: Let the Agent install via SkillHub instructions

The installation prompt provided by the SkillHub skill page is: first read SkillHub Installation and Priority Source Configuration, then install the coordinates @user_525a1b37/fundx. You can send the following segment to your current AI Client:

Please install `@user_525a1b37/fundx` according to https://skillhub.cn/install/skillhub.md.

Method 2: Use the skillhub CLI

If skillhub is already available on your machine, you can install it directly into the current Agent’s skills directory. SkillHub documentation requires using --dir to point to the corresponding directory during installation; otherwise, defaulting to ./skills/ may not be recognized by the Client.

  1. Check the CLI:
command -v skillhub && skillhub --version
  1. If not installed (complete installation example, requires internet connection):
curl -fsSL https://skillhub-1388575217.cos.ap-guangzhou.myqcloud.com/install/install.sh | bash
  1. Install this Skill (replace <skills directory> with the path in your current environment):
skillhub install user_525a1b37/fundx --dir <skills directory>

Skills directories for common Clients (excerpted from SkillHub installation documentation):

  • Cursor: ~/.cursor/skills/
  • Claude Code: ~/.claude/skills/
  • Codex: ~/.codex/skills/ or .agents/skills/ under the project

After installation, refresh or restart according to each Client’s requirements to ensure the new Skill is loaded. Differences between Cursor / Codex CLI / Claude Code are mainly reflected in the skills directory location; the installation command itself follows skillhub install … --dir.

Typical Usage

Publicly available materials do not provide separate command-line subcommand examples; the usage aligns with the introduction. You can operate within the conversation after enabling the Skill:

  1. Confirm that the Agent has loaded @user_525a1b37/fundx (or that the Skill exists in the local skills directory).
  2. Provide the mutual fund code to be diagnosed (the format follows the Skill’s runtime requirements, generally the standard exchange code).
  3. Ask the Agent to invoke the Skill to complete: automatic data fetching → multi-dimensional evaluation → target audience profile, and pay attention to the exclusion notes for risky products.
  4. When using the conclusions for research or internal reference, keep the official boundary in mind: no prediction of future performance, and the output cannot replace compliance-compliant promotion or personalized investment advice from licensed institutions.

If you need to verify the version, please refer to the v2.9.15 displayed on the SkillHub page.

Use Cases and Notes

Who is it for

  • Those who need repeatable, structured diagnosis of mutual funds within an Agent workflow, rather than rewriting prompts every time.
  • Those who want to see evaluation dimensions, audience profiles, and risk exclusion perspectives simultaneously, and accept the product setting of “no performance prediction.”
  • Users of Clients that support SkillHub skills (such as Cursor, Claude Code, Codex) and can configure the skills directory according to documentation.

Limitations and Notes

  • Data sources and specific evaluation dimensions are not listed item-by-item on the crawled page; the actual output depth depends on the installed Skill content.
  • This is a community-sourced Skill; the upgrade and maintenance rhythm depends on the publisher. It is recommended to verify the version number on SkillHub before making major decisions.
  • Fund investment involves risks; this Skill is used for information organization and risk screening perspectives and does not constitute investment advice.

Conclusion

“Public Fund Elevated Diagnosis” consolidates “Code → Data → Multi-dimensional Evaluation → Audience Profile → Risk Exclusion” into a single Skill and explicitly avoids performance prediction, making it suitable for research assistance regarding mutual funds within an Agent. Installation and coordinates are subject to SkillHub:

  • SkillHub Skill Page: https://www.skillhub.cn/skills/user_525a1b37/fundx
  • Official API Address (referenced by page): https://api.skillhub.cn/user_525a1b37/fundx