broker-research-analyst

券商研报分析助手，聚焦"研报获取→结构化提取→多机构观点对比→风险识别→决策辅助"。使用时机：汇总个股券商研报、追踪行业研报、对比多机构观点、识别卖方利益冲突、验证研报时效性。默认数据源为东方财富研报中心公开接口（无需 API Key）。

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

65%

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

The body is highly actionable with concrete, verified commands and a clear pipeline, but it is somewhat verbose and suffers from broken file references plus unreferenced bundle files that weaken navigation and conciseness.

Suggestions

Tighten the verbose PDF-parsing/image-extraction section and move version constraints (Python 3.10–3.13) and accuracy figures (98.7%) into a reference document to improve conciseness.

Make the quality-gate feedback loop explicit in the body (report fails gate → drop or re-fetch → re-validate) instead of delegating it implicitly to scripts, to raise workflow clarity.

Fix broken references (agents/*.md, ../PDF_IMAGE_RESEARCH.md, config/settings.json, .env.example, requirements.txt) and add explicit links to the existing bundle files (references/机构白名单.md, references/评级体系说明.md, assets/研报分析模板.md) so progressive-disclosure navigation works.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient, but the PDF-parsing/image-extraction section is padded with detail ('准确率 98.7%', 'Python 3.10-3.13 + GB 级模型 + 推荐 GPU') and inline tool backstory that could be tightened or moved to a reference; inline version numbers should penalize conciseness per the guidelines.	2 / 3
Actionability	Provides copy-paste-ready, verified commands such as 'python report_router.py stock --code 600519 --name 贵州茅台 --days 90 --output ./cache/600519_研报分析.md' and a scripts table whose entries all exist on disk.	3 / 3
Workflow Clarity	The numbered multi-expert pipeline (1–7) and typical call chain give a clear sequence, and a quality gate exists, but the validate→fix→retry feedback loop is only implied by delegating to report_quality_gate.py rather than stated explicitly.	2 / 3
Progressive Disclosure	Structure exists and the scripts/ table is accurate, but the body references non-existent paths (agents/*.md, ../PDF_IMAGE_RESEARCH.md, config/settings.json, .env.example, requirements.txt) while the actual bundle files (references/机构白名单.md, references/评级体系说明.md, assets/研报分析模板.md) are never linked, so navigation is broken and references are poorly signaled.	2 / 3
	Total	9 / 12 Passed

Description

100%

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 specific, third-person, and answers both what and when with an explicit use-when clause and natural trigger terms. It is a strong, distinctive description.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions via the pipeline '研报获取→结构化提取→多机构观点对比→风险识别→决策辅助', matching the anchor for several specific actions rather than vague language.	3 / 3
Completeness	It states both what (the 获取→提取→对比→识别→辅助 pipeline) and when (explicit '使用时机' triggers), satisfying the highest anchor for both what and when.	3 / 3
Trigger Term Quality	The '使用时机' clause surfaces natural user phrasings such as '汇总个股券商研报', '追踪行业研报', and '对比多机构观点', giving good coverage of terms a user would actually say.	3 / 3
Distinctiveness Conflict Risk	The niche is clear (券商研报) with a named data source (东方财富研报中心公开接口) and distinct triggers, making overlap with other skills unlikely.	3 / 3
	Total	12 / 12 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: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: charliedream1/ai_quant_trade
Path: egs_skill/broker-research-analyst/SKILL.md
Commit: 3dba332

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.