chanlun

基于缠论（缠中说禅）的形态识别引擎，使用czsc库自动检测K线分型、笔、中枢，并生成一买/一卖/二买/二卖/三买/三卖等买卖点信号。支持多周期分析和形态分类（3/5/7/9/11笔形态）。

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./agent/src/skills/chanlun/SKILL.md

Quality

Content

68%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.

A well-organized, mostly lean body with executable quick-start code and verified one-level-deep references for core concepts and buy/sell points. Weaknesses center on the implicit user workflow (no sequenced steps or validation checkpoints) and signal functions that are listed but never demonstrated.

Suggestions

Add a short numbered workflow section (prepare bars → construct CZSC → inspect bi_list/中枢 → apply a signal function), including a validation checkpoint such as confirming a minimum bar count before interpreting signals.

Show one executable example of calling a signal function (e.g. cxt_first_buy_V221126 on a CZSC/Cat object) and reading its 1/-1/0 output, since the functions are currently only named.

Add a reference file for the full signal-function catalog and multi-timeframe analysis so the remaining ~28 functions and 多周期分析 have a natural home in the bundle.

Dimension	Reasoning	Score
Conciseness	The body is lean and table-driven, and the niche Chan-theory terminology justifies its brief definitions, but time-sensitive details — the 'v0.9.68+' version pin and the '43 个缠论信号函数' count — will go stale and are not in a deprecated/old-patterns section.	4 / 5
Actionability	The quick start is copy-paste ready (complete RawBar construction, CZSC usage, pip install) and the data-format table is concrete, but the 15 listed signal functions are never shown being called, leaving a minor execution gap.	4 / 5
Workflow Clarity	The stated pipeline '原始K线 → 去包含处理 → 分型识别 → 笔检测 → 中枢构建 → 买卖点判定' describes the library's internals, not the user's workflow; the actual sequence (install → prepare bars → construct CZSC → access results → apply signals) is only implied by section order with no validation checkpoints such as verifying a non-empty bi_list or minimum bar count.	3 / 5
Progressive Disclosure	All six referenced files (references/核心概念/分型.md, 笔.md, 中枢.md and references/买卖点/一买一卖.md, 二买二卖.md, 三买三卖.md) exist one level deep and are clearly signaled via the '详细文档' table columns; minor gaps include no reference covering the remaining ~28 signal functions or 多周期分析.	4 / 5
	Total	15 / 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.

A specific, comprehensive, well-differentiated description that concretely names the detection pipeline and all six buy/sell signal types. Its main weakness is the absence of any 'use when' trigger guidance, which also caps completeness.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user mentions 缠论、缠中说禅、分型/笔/中枢、买卖点信号, or asks to analyze K-line structure on A股/crypto/ futures data.'

Include common synonyms such as '背驰' and an anglicization ('Chan theory / chanlun') so users phrasing the request in other ways still match.

Mention the supported markets (A股、加密货币、期货) in the description itself, since users often phrase requests by market rather than by theory.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete, comprehensive actions — '自动检测K线分型、笔、中枢', '生成一买/一卖/二买/二卖/三买/三卖等买卖点信号', '支持多周期分析和形态分类（3/5/7/9/11笔形态）' — in third-person voice, covering the engine's full capability set.	5 / 5
Completeness	The 'what' is clear and detailed (detection pipeline plus signal generation), but there is no 'use when' clause or equivalent trigger guidance — guidelines explicitly cap completeness at 3 for this.	3 / 5
Trigger Term Quality	Strong natural keywords a Chan-theory user would say ('缠论', '缠中说禅', '分型', '笔', '中枢', '买卖点', '一买', 'czsc'), but a few common variations like '背驰' and any English rendering ('Chan theory') are missing.	4 / 5
Distinctiveness Conflict Risk	The description occupies a clear niche — Chan-theory (缠论) pattern recognition via the czsc library — with distinct domain triggers and minimal overlap risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 6 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/chanlun/SKILL.md
Commit: 251b094

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.