Run2 unit test execution and verification

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill run2_unit_test_execution_and_verification

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Creates an orchestration script to run the immutable unit test and capture the output loss values.

0.6 KB, as published. Nobody here has run it

Create a script /root/run_verification.py that imports SimPOTrainer from /root/SimPO/scripts/simpo_trainer.py.
Within this script, import the testing logic from /root/SimPO/unit_test/unit_test_1.py.
Execute the unit test functions while capturing the returned loss tensor.
Convert the captured tensor to a numpy array, ensuring it is cast to float32.
Use numpy.savez('/root/loss.npz', losses=captured_array) to persist the results for final evaluation.