CtrlK
BlogDocsLog inGet started
Tessl Logo

sql-analysis

Answer a quantitative business question by writing a SQL query against the data warehouse, validating it, and presenting the result. Use when the user asks "how many...", "what's the trend of...", "compare X vs Y over...", "what's our top N...", or anything that resolves to a query against tabular data. Produces a small result table plus the underlying query.

70

Quality

88%

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

88%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 tight, well-sequenced SOP: it turns a business question into a metric, drafts SQL under explicit conventions, validates before reporting with a genuine fix-first feedback loop, and mandates a consistent report structure with anti-patterns and delegation rules. The main gap is execution mechanics — it never says how to actually run a query against the warehouse.

Suggestions

In step 4, name the concrete execution mechanism (e.g. the warehouse CLI, SQL tool, or connection command) so 'Run the query' is directly actionable rather than assumed.

Add one short worked example (question → restated metric → final report) to anchor the template in a concrete case; keep it under 15 lines to protect conciseness.

If the skill grows, move the report template and validation checklist into a reference file and keep SKILL.md as the overview.

DimensionReasoningScore

Conciseness

Lean and efficient throughout — no explanation of what SQL or a warehouse is, no padding; every line ('Comment any non-obvious filter', 'A surprising row count is almost always a bug') adds non-obvious guidance that earns its place.

5 / 5

Actionability

Concrete, executable guidance: the restatement template '<metric> by <grouping> over <time window>, filtered by <filter>', specific SQL conventions (CTEs, explicit column lists, WHERE-clause windows), a copy-paste report skeleton, and named delegation targets. Not a 5 because step 4 says 'Run the query' without ever naming how the query is executed (which tool/CLI/connection against the warehouse).

4 / 5

Workflow Clarity

Clear 5-step sequence with an explicit validation checkpoint and feedback loop: row-count, NULL, and numeric sanity checks, with 'If anything looks off, do not report the number — fix the query first' — a direct match for the top anchor.

5 / 5

Progressive Disclosure

Well-organized sections (Steps, Anti-patterns, When to delegate) with no buried or nested references — but no bundle files exist, and the ~70-line body slightly exceeds the under-50-line simple-skill case where organization alone earns a 5; the report template could live in a reference if the skill grows.

4 / 5

Total

18

/

20

Passed

Description

83%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 domain and concrete actions, gives an explicit 'Use when...' clause with natural trigger phrases users would say, and states the output format. The only weaknesses are minor — slightly broad 'anything that resolves to a query against tabular data' and a few missing common synonyms like 'database' or 'SQL report'.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'writing a SQL query against the data warehouse, validating it, and presenting the result' plus 'Produces a small result table plus the underlying query'. Not a 5 because coverage has minor gaps: no mention of which SQL engine/dialect or what form validation takes.

4 / 5

Completeness

Clearly answers both 'what' (write a SQL query against the data warehouse, validate it, present the result) and 'when' ('Use when the user asks "how many...", ...') with concrete trigger phrases — a direct match for the 5 anchor.

5 / 5

Trigger Term Quality

Strong natural trigger phrases users would actually say: '"how many..."', '"what's the trend of..."', '"compare X vs Y over..."', '"what's our top N..."', 'query against tabular data'. Not a 5 because common variations like 'SQL', 'database', 'sum/total of X', or specific metric phrasings are not covered.

4 / 5

Distinctiveness Conflict Risk

Clear niche (quantitative business questions → validated SQL against the data warehouse) with distinct triggers, but 'anything that resolves to a query against tabular data' plus 'compare X vs Y over...' leaves minor overlap risk with charting/dataviz or general data-analysis skills.

4 / 5

Total

17

/

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
agentscope-ai/agentscope-java
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.