CtrlK
BlogDocsLog inGet started
Tessl Logo

smart-ask-data

问数端到端编排(native CLI 版):从候选 KN 选定知识网络,用 bkn object-type 发现对象类与字段, 由编排层 LLM 生成 SQL,再由 ontology dataview query 执行取数; 最后输出中文结论与口径说明。 当用户需要指标、统计、趋势、SQL 取数或数据查询时使用。

68

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

Well-structured content with a clear sequenced workflow, concrete command forms, and excellent progressive disclosure into real reference files. The main weakness is repeated restatement of the delegation/no-token constraints across sections, which could be consolidated for token efficiency.

Suggestions

Consolidate the '委托 ontology-core / 无须 token / Never 直接执行 CLI' reminders into a single stated-once constraint block to reduce repetition across 调用方式, 子技能依赖, 委托命令形态, 与 smart-data-analysis 的关系, and 配置 sections.

Surface a brief explicit validate→proceed checkpoint inline in the 主流程 checklist (e.g. '未拿到 meta_table_name → 回 step 2.4 补齐') rather than only in references/sql-execute.md, so the gating loop is visible without following a link.

Tighten the 步骤约束 summary and 注意事项, which partially overlap (e.g. no-fabrication rules appear in both '总结' constraint #7 area and the 注意事项 list).

DimensionReasoningScore

Conciseness

Dense and information-rich, assuming domain competence, but the delegation model ('Never 直接执行 ontology CLI', '无须 token', '委托 ontology-core') is restated across several sections and could be trimmed.

4 / 5

Actionability

Provides concrete command forms — 'bkn object-type list <kn-id>', 'dataview query <dataview-id> --sql "..."', 'ontology --user-id <accountId> <command>' — and a copyable step checklist; placeholders are appropriate given the delegation model, with only minor gaps.

4 / 5

Workflow Clarity

Clear 5-step sequence with a progress checklist, explicit 3a/3b routing, and gating constraints ('Schema 发现先于取数', '禁止混用', '只允许 SELECT/WITH'); the detailed validate→fix→retry loop lives in references rather than the body, leaving minor checkpoint gaps.

4 / 5

Progressive Disclosure

Body is an orchestration overview that maps each step to a one-level-deep reference (kn-resolve.md, schema-discovery.md, sql-execute.md, tool-examples.md), all of which exist and are clearly signaled in a '必读 references' table.

5 / 5

Total

17

/

20

Passed

Description

88%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 strong, specific description that names concrete pipeline actions and tooling and provides an explicit 'use when' clause with natural trigger terms. Minor overlap risk with sibling analytics skills and a few missing trigger synonyms keep it just below perfect.

DimensionReasoningScore

Specificity

Lists multiple concrete actions with named tooling — '从候选 KN 选定知识网络', '用 bkn object-type 发现对象类与字段', '由编排层 LLM 生成 SQL', 'ontology dataview query 执行取数', '输出中文结论与口径说明' — giving comprehensive coverage of the pipeline.

5 / 5

Completeness

Explicitly answers both 'what' (the end-to-end orchestration steps) and 'when' ('当用户需要指标、统计、趋势、SQL 取数或数据查询时使用') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Natural trigger terms '指标、统计、趋势、SQL 取数或数据查询' cover the common phrasings a user would say, with a couple of synonyms (SQL 取数 / 数据查询), though a few natural variants like 报表/数据分析 are absent.

4 / 5

Distinctiveness Conflict Risk

The niche is fairly distinct (ontology native CLI 问数 via bkn/dataview), but the broad triggers (统计/趋势/数据查询) create minor overlap risk with the sibling smart-data-analysis skill.

4 / 5

Total

18

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 1 missing, 6 suspicious

Warning

Total

14

/

16

Passed

Repository
UnicomAI/wanwu
Reviewed

Table of Contents

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.