CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

answering-natural-language-questions-with-dbt

Writes and executes SQL queries against the data warehouse using dbt's Semantic Layer or ad-hoc SQL to answer business questions. Use when a user asks about analytics, metrics, KPIs, or data (e.g., "What were total sales last quarter?", "Show me top customers by revenue"). NOT for validating, testing, or building dbt models during development.

Quality

86%

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

The canonical home for this skill is dbt-labs/dbt-agent-skills

SKILL.md
Quality
Evals
Security

Quality

Content

81%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 well-crafted operational skill: the four-approach escalation path is unambiguous, failure branches are explicit, and guidance is delivered through dense tables and concrete commands rather than prose. The main deductions are the redundancy between the mermaid flowchart, Quick Reference table, and overlapping red-flag/mistake tables, and SQL examples that are placeholders rather than complete.

DimensionReasoningScore

Conciseness

The body is dense and table-driven ("Rationalizations to Resist", "Red Flags - STOP", "Common Mistakes") with no explanatory padding, but the ~45-line mermaid flowchart substantially duplicates the Quick Reference table and the four Approach sections, and the Rationalizations/Mistakes tables overlap — trimming would land it at anchor 5.

4 / 5

Actionability

Concrete tool sequences ("list_metrics", "get_metrics_compiled_sql", then "execute_sql") and executable jq commands ("jq '.nodes | to_entries | map(select(...))' target/manifest.json") are given, but several SQL snippets are placeholders ("-- ... compiled metric logic", "SELECT * FROM (compiled_metric_sql)", "GROUP BY ...") rather than complete examples, keeping it below anchor 5.

4 / 5

Workflow Clarity

The multi-step process is clearly sequenced via the priority table and numbered approaches, with explicit failure branches and feedback loops ("try_compiled -->|fail| check_discovery", "Always exhaust options before saying 'cannot answer'") that constitute error-recovery checkpoints — a clear anchor 5 match with no validation gaps for this read/execute workflow.

5 / 5

Progressive Disclosure

The single file is well-organized into clearly headed, navigable sections that each earn their place, with no nested or buried references (no bundle files exist). It sits at anchor 4 rather than 5 only because at ~190 lines some content (e.g., the full decision flowchart or detailed SQL modification patterns) could be split into one-level-deep reference files.

4 / 5

Total

17

/

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: third-person, concrete about what it does and when to use it, with natural trigger phrases and example questions, and an explicit NOT-for clause that cleanly separates it from dbt development skills. The only minor gaps are a couple of missing natural trigger synonyms (e.g., reporting, dashboard) and action coverage that stops just short of comprehensive.

DimensionReasoningScore

Specificity

"Writes and executes SQL queries against the data warehouse using dbt's Semantic Layer or ad-hoc SQL" names the domain and concrete actions with specific tooling, going beyond the 1-2 generic actions of anchor 3, though it stops short of the comprehensive multi-action coverage of anchor 5.

4 / 5

Completeness

Both what ("Writes and executes SQL queries... to answer business questions") and when ("Use when a user asks about analytics, metrics, KPIs, or data") are explicitly answered with concrete trigger phrases, and the NOT-for clause adds explicit scope boundaries — a clear anchor 5 match, not merely the adequate-but-could-be-more-explicit anchor 4.

5 / 5

Trigger Term Quality

"analytics, metrics, KPIs, or data" plus natural example phrases like "What were total sales last quarter?" and "Show me top customers by revenue" give good keyword coverage, but common variations such as "reporting", "dashboard", or "revenue breakdown" are absent.

4 / 5

Distinctiveness Conflict Risk

The dbt Semantic Layer / data warehouse niche with distinct triggers, plus the explicit exclusion "NOT for validating, testing, or building dbt models during development", gives minimal conflict risk with other dbt or SQL skills.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

dbt-labs/dbt-agent-skills

Path

skills/dbt/skills/answering-natural-language-questions-with-dbt/SKILL.md

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.