ashare-pre-st-filter

A 股 ST/*ST 风险预测框架 — 基于最新中报/三季报或业绩预告/快报，预测下一财年是否会因营收、利润、净资产、分红不达标而被风险警示，并将新浪监管处罚记录作为独立证据面纳入风险等级。仅适用于 A 股，不预测财务造假。

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./agent/src/skills/ashare-pre-st-filter/SKILL.md

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a dense, highly actionable operational workflow with concrete interfaces, formulas, a bundled script, and pseudocode. Its weaknesses are token duplication (interface list and prohibitions restate earlier sections) and limited offloading of reference material to separate files.

Suggestions

Remove or merge the standalone "依赖的 tushare 接口清单" table since the data-mapping table already covers the same interfaces with richer detail.

Tighten the "禁止事项" list to cross-reference the relevant R-sections rather than re-deriving each rule in full.

Consider moving the long threshold tables and the full end-to-end pseudocode into a reference file linked from SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly efficient domain-operational content, but the standalone "依赖的 tushare 接口清单" duplicates the data-mapping table and the "禁止事项" section restates rules already detailed in R3/R4, adding avoidable tokens.	3 / 5
Actionability	Fully executable: concrete tushare/akshare interface names with parameters and time windows, an exact量纲公式 (cash_div_tax / 10 × total_share), a real bundled script with copy-paste CLI invocations, and complete end-to-end pseudocode.	5 / 5
Workflow Clarity	M0 through Step 6 are clearly sequenced with data-source fallbacks, evidence-priority overrides, and confidence-demotion rules; only minor validation/checkpoint gaps prevent a 5.	4 / 5
Progressive Disclosure	Well-organized into clearly headed sections with a real, clearly signaled bundle reference (scripts/fetch_sina_penalties.py, verified to exist); the main gap is the duplicated interface list and reference-grade tables inlined in the body.	4 / 5
	Total	16 / 20 Passed

Description

75%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is highly specific and distinctive with a clear niche, but it omits an explicit "Use when..." trigger clause, which caps its completeness. Trigger-term coverage is strong but missing a couple of natural synonyms.

Suggestions

Add an explicit "Use when..." trigger clause (e.g., "Use when 分析 A 股个股的 ST/*ST 或退市风险") to raise completeness above 3.

Include the synonym 退市风险 / 退市 in the description to improve trigger-term coverage toward 5.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete, comprehensive actions — predicting next-fiscal-year ST/*ST triggered by revenue, profit, net assets, and dividend non-compliance plus incorporating Sina penalty records as an independent evidence face.	5 / 5
Completeness	It clearly states what the skill does but lacks any "Use when..." or equivalent explicit trigger guidance; per the rubric a missing trigger clause caps completeness at 3.	3 / 5
Trigger Term Quality	Good natural-term coverage ("ST/*ST 风险预测", "风险警示", the four metric names), but the description omits common synonyms like 退市风险 that the body's triggers actually use.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (A-share ST/*ST risk prediction) with explicit exclusions ("仅适用于 A 股，不预测财务造假"), giving minimal conflict risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (515 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing, 2 deeper-than-1-level, 4 suspicious	Warning

	Total	13 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/ashare-pre-st-filter/SKILL.md
Commit: d71ed8d

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.