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

73%

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 ./configs/microservice/bff-service/configs/agent-skills/ontology/smart-data-analysis/SKILL.md

Quality

Content

71%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 orchestration playbook with explicit sequenced workflows, validation checkpoints, executable CLI templates, and clear one-level-deep references to sibling skills. Its main weakness is repetition: several governance rules (no direct CLI, no cross-fallback) are restated many times, adding tokens without new information.

Suggestions

Consolidate the repeated 'Never execute ontology CLI directly — delegate to ontology-core' rule to a single authoritative statement; the 5+ restatements across sections are redundant token cost.

State the '禁止交叉兜底' rule once in its dedicated section and reference it by name elsewhere rather than re-explaining the full rationale in the 总入口原则 list.

Move the detailed JSON condition-wrapper example and the SQL-vs-query-vs-properties comparison table toward a referenced reference file (e.g. ontology-core/bkn.md) so the SKILL.md body stays a lean overview.

Dimension	Reasoning	Score
Conciseness	The ~270-line body is densely packed with routing rules, tables, and checklists that are largely necessary, but it repeats the same constraints many times (the 'Never execute ontology CLI directly / delegate to ontology-core' rule is restated 5+ times, and the no-cross-fallback rule is stated then re-stated in a dedicated section), which is noticeable padding for an orchestrator skill.	3 / 5
Actionability	It provides concrete, executable guidance — explicit CLI commands (bkn object-type query/get, dataview query --sql), a correct JSON body template with the required 'condition' wrapper, an explicit wrong-example error message, and concrete handoff field tables — with only minor gaps (no full copy-paste SQL example, ot-id selection left to LLM judgment).	4 / 5
Workflow Clarity	Multi-step processes are explicitly sequenced with copy-and-tick checklists for both 找表 and 问数 branches, a top-level 5-step orchestration flow, explicit validation/feedback checkpoints (占位未填 → 告知用户并停止; schema discovery before SQL; re-clarify on ambiguity), and explicit error-recovery guidance for when a branch fails.	5 / 5
Progressive Disclosure	Structure is good with clear section headers and one-level-deep references to sibling skills ([smart-search-tables/SKILL.md], [smart-ask-data/SKILL.md], [ontology-core/SKILL.md]) and to ontology-core's references/*.md (尤其 bkn.md); no references/ scripts/ assets/ bundle files exist in this skill's own directory, so there is no deep nesting, though some reference-target detail is inlined that could live in those referenced files.	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 a detailed, third-person statement that concretely defines the orchestrator role and its boundaries with sibling skills, and includes an explicit 'use when' trigger. It is dense with internal jargon (kn, KN, ontology) which slightly weakens natural trigger phrasing and breadth of plain-language synonyms.

Suggestions

Add plain-language trigger synonyms a user would actually say (e.g. '查数据、算指标、找表、看报表') alongside the internal kn/KN terminology to improve natural trigger coverage.

Tighten the 'when' clause into a short standalone sentence ('Use when the user asks any data question or needs to switch between business knowledge networks') so the trigger reads as a crisp signal rather than being embedded in a long run-on.

Trim the internal routing detail (delegate to ontology-core, no direct CLI) from the description — it is governance, not capability, and slightly dilutes the what/when signal.

Dimension	Reasoning	Score
Specificity	Names the domain (data analysis orchestration) and several concrete actions — kn 分域、上下文注入、多候选 KN 时的 LLM 决策、SQL 生成、子流程交接 — giving broad but not fully comprehensive coverage of the orchestrator's responsibilities.	4 / 5
Completeness	It clearly states 'what' (编排与路由, SQL 生成, 上下文注入) and 'when' (当用户提出任何数据类自然语言任务…时使用), satisfying both halves, though the 'when' clause is bundled into a long sentence rather than a crisp standalone trigger.	4 / 5
Trigger Term Quality	Includes natural trigger phrasing like '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' covering most user-facing scenarios, though it leans on internal jargon (kn, KN, ontology) more than plain synonyms a user would say.	4 / 5
Distinctiveness Conflict Risk	The '唯一总入口…必须先经本 skill' framing carves a clear niche as the data-stack orchestrator with explicit sibling skills (smart-search-tables / smart-ask-data / ontology-core), leaving only minor overlap risk with those closely related skills.	4 / 5
	Total	16 / 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: 99df6e4

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.