Add Resources Testing Framework documentation to check and transform skills#18
Add Resources Testing Framework documentation to check and transform skills#18dgarros wants to merge 1 commit into
Conversation
…skills Document Infrahub's YAML-driven pytest testing framework across shared common rules and skill-specific rules for checks and transforms. Elevate testing from LOW to HIGH priority and add complete test YAML examples with fixtures. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
| 4. **Caching (caching-)** -- MEDIUM. Display label caching with parent relationships, batch loading timing issues, no-op mutation workarounds. | ||
|
|
||
| 5. **Testing (testing-)** -- HIGH. Resources Testing Framework, YAML-driven pytest tests, smoke/unit/integration test kinds, always-create-tests recommendation. |
There was a problem hiding this comment.
Pedantic, but there is a difference in width on these lines from the other items. Should probably be made consistent.
|
|
||
| Transforms should always have accompanying test definitions. The Resources Testing Framework provides test kinds for both Python and Jinja2 transforms, from lightweight syntax validation to full integration testing. | ||
|
|
||
| See [Resources Testing Framework](../../common/rules/testing-resource-framework.md) for the general framework overview, YAML structure, and how to run tests. |
There was a problem hiding this comment.
Do we want to be a little more specific about what kind of tests must (or should) be created for transforms? For example, a transform should create a specific output, not a fuzzy output, so we might want to be specific about creating unit tests (and not just smoke tests) for transforms. The instructions are a little ambiguous about which kind of tests to use for a transform definition, and the LLM might be a bit lazy and just create a smoke test instead of a unit test. Similarly, an integration test might be a step too far in testing for the majority of transform definitions. Integration tests take additional resources, and a unit test is probably sufficient.
3 participants
Document Infrahub's YAML-driven pytest testing framework across shared common rules and skill-specific rules for checks and transforms. Elevate testing from LOW to HIGH priority and add complete test YAML examples with fixtures.