dbt-labs/dbt-agent-skills
A curated collection of Agent Skills for working with dbt, to help AI agents understand and execute dbt workflows more effectively.
91
Does it follow best practices?
Validation for skill structure
migrating-dbt-project-across-platforms
skills/dbt-migration/skills/migrating-dbt-project-across-platforms/SKILL.md
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-structured description with explicit 'Use when' guidance and strong trigger terms covering specific platforms and migration scenarios. The main weakness is that it could be more specific about the concrete actions performed during migration (e.g., syntax conversion, function mapping, data type transformation) rather than the general 'identify and fix SQL dialect differences'.
Suggestions
Expand the capabilities section to list specific concrete actions like 'converts SQL syntax, maps platform-specific functions, transforms data types, updates configuration files'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (dbt project migration between data platforms) and mentions one action (identify and fix SQL dialect differences), but doesn't list multiple concrete actions like 'convert syntax', 'update functions', 'transform data types'. | 2 / 3 |
Completeness | Explicitly answers both what (migrating dbt projects using dbt Fusion's real-time compilation to identify and fix SQL dialect differences) and when (starts with 'Use when migrating a dbt project from one data platform to another'). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'migrating', 'dbt project', 'data platform', 'data warehouse', 'Snowflake', 'Databricks', 'SQL dialect', and 'dbt Fusion' - good coverage of terms a user would naturally mention. | 3 / 3 |
Distinctiveness Conflict Risk | Very specific niche combining dbt, cross-platform migration, and dbt Fusion tooling - unlikely to conflict with general SQL skills, generic dbt skills, or single-platform skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-structured migration skill with excellent workflow clarity and actionability. The 7-step process with explicit validation checkpoints and a progress checklist provides clear guidance for a complex multi-step operation. Minor verbosity in the introductory sections and some redundancy between the 'Don't Do These Things' section and earlier content could be trimmed for better token efficiency.
Suggestions
Trim the 'core approach' paragraph in the introduction - the workflow itself demonstrates the approach without needing to explain it upfront
Consolidate redundant guidance between the workflow steps and 'Don't Do These Things' section (e.g., 'don't pre-fix issues' is stated multiple times)
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some unnecessary explanation (e.g., explaining what Fusion does, the 'core approach' section restates concepts). The 'Don't Do These Things' section has some redundancy with earlier content. Could be tightened by ~20%. | 2 / 3 |
Actionability | Provides concrete, executable commands throughout (dbtf compile, rm -rf target/, dbt test --select test_type:unit). Clear bash examples, specific file paths, and copy-paste ready commands. The workflow steps are specific and actionable. | 3 / 3 |
Workflow Clarity | Excellent 7-step workflow with explicit validation checkpoints (compile → fix → recompile loop). Includes a copyable progress checklist, clear success criteria upfront, and explicit feedback loops for error recovery. Validation gates are clearly marked before expensive operations. | 3 / 3 |
Progressive Disclosure | Well-structured with clear table of contents, one-level-deep references to detailed docs (installing-dbt-fusion.md, generating-unit-tests.md, switching-targets.md). Main content stays focused on the workflow while appropriately linking to reference materials. | 3 / 3 |
Total | 11 / 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 |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |