dbt-transformation-patterns
Master dbt (data build tool) for analytics engineering with model organization, testing, documentation, and incremental strategies. Use when building data transformations, creating data models, or implementing analytics engineering best practices.
74
62%
Does it follow best practices?
Impact
95%
1.20xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/data-engineering/skills/dbt-transformation-patterns/SKILL.mdQuality
Discovery
67%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 competent with a clear 'Use when' clause and identifies the dbt domain well. However, it leans toward category-level language rather than listing concrete specific actions, and the trigger terms could be expanded to include more natural user phrases and dbt-specific terminology. The word 'Master' at the beginning is slightly promotional but doesn't significantly detract.
Suggestions
Replace category labels with more concrete actions, e.g., 'write staging and mart models, configure materializations, define schema tests, set up sources and refs, generate documentation'
Add more natural trigger terms users would say, such as 'SQL models', 'Jinja macros', 'dbt run/test', 'materializations', 'ref()', 'sources.yml', or 'warehouse transformations'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (dbt/analytics engineering) and mentions some actions like 'model organization, testing, documentation, and incremental strategies,' but these are more like category labels than concrete specific actions. Compare to a score-3 example which would list things like 'create staging models, write schema tests, configure incremental materializations, generate dbt docs.' | 2 / 3 |
Completeness | Clearly answers both 'what' (model organization, testing, documentation, incremental strategies) and 'when' with an explicit 'Use when...' clause covering building data transformations, creating data models, or implementing analytics engineering best practices. | 3 / 3 |
Trigger Term Quality | Includes relevant terms like 'dbt', 'data build tool', 'data transformations', 'data models', and 'analytics engineering', which are good. However, it misses common natural variations users might say such as 'SQL models', 'ref()', 'sources', 'materializations', 'dbt run', 'dbt test', 'Jinja', 'warehouse transformations', or '.yml schema files'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'dbt' and 'analytics engineering' provides a fairly distinct niche, but phrases like 'data transformations' and 'creating data models' are broad enough to potentially overlap with general SQL skills, ETL tools, or other data pipeline skills. The term 'data build tool' helps but the broader language introduces some conflict risk. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, production-ready dbt patterns with excellent executable code examples covering the full model lifecycle. Its main weaknesses are its monolithic structure (everything inline rather than progressively disclosed across files) and the lack of explicit validation workflows — there's no guidance on when to run tests, how to verify incremental models are working correctly, or how to recover from common failures. The content would benefit significantly from being restructured into an overview with references to detailed pattern files.
Suggestions
Restructure into a concise SKILL.md overview with references to separate files like PATTERNS.md, INCREMENTAL.md, and COMMANDS.md to improve progressive disclosure
Add an explicit end-to-end workflow with validation checkpoints, e.g., '1. Create staging model → 2. Run dbt run --select stg_model → 3. Run dbt test --select stg_model → 4. If tests fail, fix and re-run → 5. Only then proceed to intermediate layer'
Remove the 'When to Use This Skill' and 'Core Concepts' sections — Claude knows what dbt layers are; jump straight to the project config and patterns
Add error recovery guidance for common failure modes like incremental model schema changes, failed freshness checks, and test failures in production
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is quite long (~400+ lines) with extensive inline code examples that could be split into referenced files. Some sections like the 'When to Use This Skill' list and 'Core Concepts' explain things Claude already knows about dbt architecture. However, the code examples themselves are lean and well-commented. | 2 / 3 |
Actionability | Excellent actionability with fully executable SQL models, complete YAML configurations, working macros, and copy-paste ready dbt commands. Every pattern includes real, runnable code with proper dbt Jinja syntax, not pseudocode. | 3 / 3 |
Workflow Clarity | The patterns are well-sequenced from sources → staging → intermediate → marts, and the commands section covers the build workflow. However, there are no explicit validation checkpoints or feedback loops — e.g., no guidance on running 'dbt test' after each layer, no error recovery steps for failed incremental runs, and no verification workflow for the overall build process. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of content with all patterns, examples, and reference material inline. The extensive code examples for 7+ patterns, command reference, and best practices should be split into separate files (e.g., PATTERNS.md, COMMANDS.md, INCREMENTAL.md) with the SKILL.md serving as a concise overview with links. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (564 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
6e3d68c
- 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.