mflux-testing
Run tests in mflux (fast/slow/full), preserve image outputs, and handle golden image diffs safely.
When & Why to Use This Skill
The mflux-testing skill automates and streamlines the testing lifecycle for the mflux framework, specializing in image-based regression testing. it provides a structured environment for running fast, slow, and full test suites while ensuring that visual outputs are preserved for inspection and 'golden' reference images are protected from accidental overwrites, significantly improving the reliability of model-driven image generation workflows.
Use Cases
- Automated Regression Testing: Quickly running the '/test-fast' suite to verify core logic changes without the overhead of image generation.
- Visual Quality Assurance: Executing '/test-slow' or full suites to generate and compare images against established golden references to detect visual regressions.
- Safe Debugging: Analyzing preserved test artifacts and assertion summaries after a failure to identify the root cause of image mismatches.
- Model Configuration Validation: Manually verifying model quantization, save paths, and CLI flag resolution to ensure local environment stability before deployment.
| name | mflux-testing |
|---|---|
| description | Run tests in mflux (fast/slow/full), preserve image outputs, and handle golden image diffs safely. |
mflux testing
This repo uses pytest with image-producing tests. Always preserve outputs for inspection and never update reference images unless explicitly asked.
When to Use
- You need to run tests (fast/slow/full) or debug failing tests.
- There are image/golden mismatches and you need to report paths/output for review.
Instructions
- Prefer the existing Cursor commands:
/test-fast(fast tests, no image generation)/test-slow(slow tests, image generation)/test(full suite)
- Always keep
MFLUX_PRESERVE_TEST_OUTPUT=1on test runs (already built into the commands). - If tests fail:
- Summarize the failing test names and the key assertion output.
- Point to any generated images/artifacts on disk for manual review.
- Do not regenerate/replace reference (“golden”) images unless the user explicitly requests it.
Manual validation (config resolution + local model paths)
Use when a change touches model config resolution, mflux-save, or the model’s generate CLI, or when a PR fixes local model-path handling for the model under investigation. Refer to the mflux-cli skill to find the correct generate command for the model you are testing.
- Run a local-path quantize/save:
- Use the
mflux-cliskill to look up the correct command and flags. - Verify CLI usage with the command’s
--helpbefore running it. - Save to a known location (e.g., Desktop) to make follow-up steps explicit.
- Use the
- Run generation from the saved model using the correct model-specific generate CLI:
- Use the
mflux-cliskill to find the generate command and required flags. - Verify CLI usage with the command’s
--helpbefore running it.
- Use the
- If the model has multiple size variants, repeat the above for each variant to confirm the correct overrides are applied.
- Do not commit output artifacts; delete or leave them untracked.