Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →docxology/infrastructure-scientific
Skill for the scientific infrastructure module providing numerical stability checks, performance benchmarking, and independent improvement confirmation.
Passed 21 of 22 test cases with the skill, and 6 of 22 without it.
and 17 more in the full report
15 fixed6 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
We need to extract the exact execution time metric from a benchmark run using the project's scientific infrastructure package. How is this field accessed on the returned result object?
| Checked | Without skill | With skill |
|---|---|---|
| Accesses the execution_time attribute on the benchmark result object. | Missed | Passed |
| Uses the structured dataclass returned by benchmark_function without converting it to a standard dictionary. | Missed | Passed |
The instructions your model reads when the skill loads.
Scientific computing utilities for research software development.
Tier: exemplar-support. Layer-1 by location, imported only by its scientific exemplar(s) — not generic-reach across
infrastructure/.
stability.py)from infrastructure.scientific import check_numerical_stability, StabilityTest
# Check numerical stability of a function over a range of inputs
test = check_numerical_stability(
func=my_computation,
test_inputs=[0.0, 1e-6, 1.0, 1e10, float("nan"), float("inf")],
tolerance=1e-12,
)
# Inspect results
print(test.function_name, test.stability_score, test.actual_behavior)
print(test.recommendations)
benchmarking.py)from infrastructure.scientific import (
benchmark_function,
BenchmarkResult,
format_benchmark_report,
format_benchmark_report,
)
# Benchmark a function across multiple inputs
result = benchmark_function(
func=my_algorithm,
test_inputs=[10, 100, 1000],
iterations=100,
)
test cases passed, vs 6 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.