The body is an exemplar of a lean orchestration skill: concrete tool calls with parameter and return-field contracts, explicit validation and retry loops, ready-to-use dialogue and prompt templates, and clear behavioral red lines. Its only weakness is inherited from the frontmatter — the trigger guidance lives here rather than in the description where it aids skill selection.

Dimension	Reasoning	Score
Conciseness	The body is lean throughout: trigger phrases, four tool invocations, field-by-field attribution semantics, dialogue templates, a prompt template, and red lines — no padding and no explanations of concepts Claude already knows. Every token earns its place.	5 / 5
Actionability	Fully concrete guidance: named tool calls with parameters (`extract_shadow_strategy(journal_path=...)` returning `shadow_id`; `render_shadow_report` returning `html_path`/`pdf_path`/`delta_pnl`), copy-ready dialogue templates with format specs (`{delta_pnl:+.0f}`), and a complete structured LLM translator prompt template with [上下文]/[任务]/[输出] sections.	5 / 5
Workflow Clarity	The four-step sequence is clearly numbered with explicit validation and feedback loops: a user-confirmation loop ("如果用户说'不像'，提高 min_support 重跑"), a hard sample-size check ("profitable roundtrips < 5 → 直接 raise"), and error recovery ("weasyprint 失败时自动降级成 HTML-only").	5 / 5
Progressive Disclosure	No bundle files exist and none are needed; all ~73 lines are runtime-relevant and well-sectioned (触发条件, workflow, output interpretation, templates, red lines), with nothing that warrants offloading to a reference file. Fits the simple-skill exception for progressive disclosure.	5 / 5
	Total	20 / 20 Passed

Description

62%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 excels at stating concrete, quantified capabilities and occupies a distinct niche, but it never tells Claude when to load the skill and lacks trigger phrasing or synonym coverage in the description itself. Moving the body's trigger phrases ("提炼我的策略" / "我的盈利模式" etc.) into the description would resolve both weaknesses at once.

Suggestions

Append an explicit trigger clause to the description, e.g. "Use when the user says '提炼我的策略' / '训练影子' / '我的盈利模式' / '我能多赚多少' or asks to backtest their own trading style."

Add natural synonym variations for discoverability (交割单/交易日记/trade journal, 回测/backtest) so users phrasing the need differently still match.

Drop the second-person tagline "叙事：你的影子，没有情绪噪音" from the description (or reword in third person) and consider mentioning the optional daily signal scan to round out capability coverage.

Dimension	Reasoning	Score
Specificity	The description lists four concrete, quantified actions — "提炼盈利模式（3-5 条人话规则）", "跨 A股/港股/美股/crypto 多市场回测", "差值归因", "8-section PDF 报告" — but omits the optional daily signal-scan capability and appends a second-person tagline ("你的影子，没有情绪噪音"), leaving minor coverage gaps that fit the 4 anchor rather than the comprehensive 5.	4 / 5
Completeness	The 'what' is explicit and concrete (extract rules, backtest across four markets, attribute deltas, render PDF), but 'when' is entirely absent from the description — trigger guidance lives only in the body — so per the guideline a missing 'Use when...' clause caps completeness at 3.	3 / 5
Trigger Term Quality	It contains relevant domain keywords a user might say ("交割单", "盈利模式", "回测") but no explicit trigger phrasing, no 'use when' guidance, and no synonyms or common variations, matching the anchor 'Some relevant keywords but missing common variations or synonyms'.	3 / 5
Distinctiveness Conflict Risk	The 'Shadow Account' niche — trade-journal-derived rule extraction with multi-market delta attribution — is highly specific with named markets and a unique output artifact (8-section PDF), giving it a clear niche with minimal conflict risk with other skills.	5 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/shadow-account/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.