smart-data-analysis

数据分析员工（Data Analyst Agent）的唯一总入口：凡与数据资产、取数、指标、表/视图、治理职责、知识网络、统计或分析相关的问题，必须先经本 skill 做编排与路由，再进入找表或问数等子流程。负责 kn 分域、上下文注入（accountId / date）、多候选 KN 时的 LLM 决策、问数分支的 SQL 生成；与 smart-search-tables / smart-ask-data / ontology-core 的交接。当用户提出任何数据类自然语言任务、或需在多条业务 KN 间切换时使用；所有 ontology CLI 执行均委托 ontology-core 完成，本 skill 不直接执行 CLI。

Quality

83%

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

Quality

Content

73%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 well-structured, actionable orchestration guide with clear sequenced workflows, validation checkpoints, and explicit failure handling. Its main weaknesses are length/repetition and the absence of a concrete SQL-generation example plus unfilled KN-id placeholders.

Suggestions

Consolidate the '禁止交叉兜底' rule into one section and reference it elsewhere rather than restating it fully in 总入口原则, its dedicated section, and 注意事项 to cut repeated tokens.

Add a concrete SELECT/WITH SQL template example (with LIMIT and schema-derived fields) in the 问数 branch to make the 'SQL 生成是本 skill 的硬责任' instruction copy-paste ready.

Move the JSON condition-wrapper example and the SQL-vs-object-type-query 判别表 into a references/ file (e.g. bkn-query.md) and link to it, improving progressive disclosure now that all detail is inline.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient domain-specific routing logic that does not explain concepts Claude already knows, but it is long and repeats the '禁止交叉兜底' rule across three sections and restates 交接契约 details, so it could be tightened per the score-3 anchor.	3 / 5
Actionability	Provides concrete checklists, explicit field names (accountId/kn_id/duty_kn_id/<ot-id>/search), and a real JSON body example with the documented error case, but lacks a concrete SELECT/WITH SQL template and leaves KN ids as placeholders, leaving minor gaps per the score-4 anchor.	4 / 5
Workflow Clarity	Sequenced checklists (编排总流程 5-step, 找表/问数 branch lists) carry explicit validation checkpoints (placeholder→stop, accountId→ask), a dedicated failure-handling section, and a short-loop feedback ('短循环找表'), matching the score-5 anchor.	5 / 5
Progressive Disclosure	Well-organized with clear section headers and clearly signaled one-level-deep references to sibling skills ([smart-search-tables], [smart-ask-data], [ontology-core]); no bundle files exist, and some inline detail (JSON body example, SQL-vs-query 判别表) could be split out, fitting the score-4 anchor.	4 / 5
	Total	16 / 20 Passed

Description

92%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 specific, complete, and distinctive, clearly stating both capabilities and trigger conditions with concrete natural-language terms. Its only minor gap is slightly less exhaustive coverage of common synonym phrasings a user might say.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'kn 分域、上下文注入（accountId / date）、多候选 KN 时的 LLM 决策、问数分支的 SQL 生成' plus handoff — giving comprehensive coverage within its data-routing domain, matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers both 'what' (orchestration/routing, KN selection, SQL generation) and 'when' via the concrete trigger clause '当用户提出任何数据类自然语言任务、或需在多条业务 KN 间切换时使用', matching the score-5 anchor.	5 / 5
Trigger Term Quality	Good coverage of natural terms users would say ('数据资产、取数、指标、表/视图、统计或分析'), but a few common phrasings/synonyms (e.g. 查数据/查表) are absent, placing it just below the comprehensive score-5 anchor.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche as the '唯一总入口' for the data-analyst role and explicitly differentiates from smart-search-tables / smart-ask-data / ontology-core, giving minimal conflict risk per the score-5 anchor.	5 / 5
	Total	19 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 16 suspicious	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: UnicomAI/wanwu
Path: configs/microservice/bff-service/configs/agent-skills/ontology/smart-data-analysis/SKILL.md
Commit: 030c20b

Reviewed: 1 day 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.