CtrlK
BlogDocsLog inGet started
Tessl Logo

database-design

设计 BK-CI 数据库表、DDL 脚本、索引、分表或数据归档策略时使用。当用户提到表结构、字段设计、索引优化、SQL 脚本命名、分片、归档迁移时优先使用。

63

Quality

79%

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

Fix and improve this skill with Tessl

tessl review fix ./ai/skills/database-design/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

61%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.

A concise, well-structured overview with good decision heuristics, but it functions as a stub: the two reference files it points to for script-naming rules and sharding strategy are missing from the bundle, and the workflow lacks explicit validation/rollback checkpoints for what is inherently risky database-change work. As written, a reader gets sound principles but no executable specifics.

Suggestions

Ship the referenced bundle files (reference/1-script-management.md and reference/2-sharding.md) or remove/inline the links — currently both references are dangling, so the script-naming and sharding detail the body promises is unreachable.

Add explicit validation checkpoints to the workflow for DDL changes: a concrete rollback step, a compatibility/gray-release verification gate, and a 'validate before applying to production' checkpoint, rather than listing rollback only as a pitfall.

Include at least one concrete example inline — e.g., the actual script naming format and version-directory layout, or a short sample DDL/migration script — so the guidance is executable even before the reference files are read.

DimensionReasoningScore

Conciseness

The ~55-line body is lean bullet-style guidance with zero padding and no explanation of concepts Claude already knows (no "what is an index" filler); every section (适用场景/不适用场景/快速指导/高信号规则/关键陷阱) earns its tokens. Matches the anchor-5 'lean and efficient' example.

5 / 5

Actionability

Guidance is directionally concrete ("先按查询路径设计索引" — design indexes from query paths first; the four table roles; the compatibility checklist) but missing key executable details: no naming-convention example, no script directory layout, no DDL example. The files that would carry those specifics (`reference/1-script-management.md`, `reference/2-sharding.md`) do not exist in the bundle, so the concrete detail is unreachable — matching anchor 3's 'some concrete guidance but incomplete'.

3 / 5

Workflow Clarity

快速指导 gives a real 7-step design sequence (service ownership → table role → indexes → scripted DDL → compatibility → scaling decision → JSON-field justification), but this is a database-change skill where the rubric caps workflow_clarity at 3 without explicit validation checkpoints. Rollback/灰度发布 appear only as a pitfall ("只改 DDL 不评估回滚、兼容和灰度发布路径") rather than as an explicit validate-before-proceed step, so the cap applies.

3 / 5

Progressive Disclosure

The body is well-sectioned and correctly delegates detail to two one-level-deep references, but neither `reference/1-script-management.md` nor `reference/2-sharding.md` exists — there is no references/ (or reference/) directory at all in the bundle, so the navigation the skill depends on is broken. Scoring against the actual bundle structure, the promised detail is unreachable, which drops it to the minimal-structure anchor 2; it is not a 1 because the SKILL.md itself is well organized and the references are clearly signaled rather than buried.

2 / 5

Total

13

/

20

Passed

Description

87%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 description: it names the niche (BK-CI database design), lists concrete design targets, and provides an explicit 'use when the user mentions...' clause with natural trigger keywords including synonyms. The only gap is limited verb variety — everything is phrased as "design X" rather than distinct actions.

DimensionReasoningScore

Specificity

The description lists several specific design targets — "设计 BK-CI 数据库表、DDL 脚本、索引、分表或数据归档策略" (design BK-CI database tables, DDL scripts, indexes, sharding or data-archiving strategies) — but all share the single verb "设计", so action variety is narrower than the comprehensive multi-verb anchor 5 (e.g., "Extract... fill... merge... convert"). It clearly exceeds anchor 3's 1–2 concrete actions.

4 / 5

Completeness

It explicitly answers both parts: what it does ("设计 BK-CI 数据库表、DDL 脚本、索引、分表或数据归档策略") and when to use it ("当用户提到...时优先使用") with a concrete list of trigger phrases — matching the anchor-5 example structure of capabilities followed by an explicit 'Use when...' clause.

5 / 5

Trigger Term Quality

"当用户提到表结构、字段设计、索引优化、SQL 脚本命名、分片、归档迁移时" gives good natural keyword coverage, including the synonym pair 分表/分片 for sharding. Not a 5 because common phrasings like 建表/数据库设计/DDL 变更 are absent; well above anchor 3's partial coverage.

4 / 5

Distinctiveness Conflict Risk

The BK-CI prefix plus a tight database-design niche (DDL scripts, sharding, archiving) with distinct triggers gives it a clear niche and minimal overlap risk with other skills; not generic or broad like the anchor-2/3 examples.

5 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
TencentBlueKing/bk-ci
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.