Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Python package creation and PyPI distribution via pyproject.toml and entry points. Use when publishing a package or setting up build configuration.
.claude/skills/athola-python-packaging/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 12% | 0% |
Modern Python packaging with pyproject.toml, uv, and best practices for distribution.
bash# Create new project with uv uv init my-package cd my-package # Add dependencies uv add requests click # Build package uv build # Publish to PyPI uv publish
Verification: Run the command with --help flag to verify availability.
instead
instead
yaml# Source layout (recommended) src/my_package/ __init__.py module.py # Flat layout (simple) my_package/ __init__.py module.py
Verification: Run the command with --help flag to verify availability.
Source layout benefits:
Minimal Project:
**Verification:** Run `pytest -v` to verify tests pass.
my-project/
├── pyproject.toml
├── README.md
├── src/
│ └── my_package/
│ └── __init__.py
└── tests/
└── test_init.pyVerification: Run pytest -v to verify tests pass.
Complete Project:
**Verification:** Run the command with `--help` flag to verify availability.
my-project/
├── pyproject.toml
├── README.md
├── LICENSE
├── .gitignore
├── src/
│ └── my_package/
│ ├── __init__.py
│ ├── cli.py
│ ├── core.py
│ └── utils.py
├── tests/
│ ├── conftest.py
│ └── test_core.py
└── docs/
└── index.mdVerification: Run pytest -v to verify tests pass.
See modules for detailed information:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | pass→pass | 16,620 | 14,827 | -11% | 1 | 1 | 0% | 1,920 | 2,143 | +12% | 0 | 0 | — |
case-03 | pass→pass | 11,749 | 8,141 | -31% | 1 | 1 | 0% | 2,042 | 2,276 | +11% | 0 | 0 | — |
case-01 | fail→pass | 12,732 | 9,019 | -29% | 1 | 1 | 0% | 2,521 | 2,576 | +2% | 0 | 0 | — |
case-04 | pass→pass | 4,622 | 3,856 | -17% | 1 | 1 | 0% | 711 | 1,440 | +103% | 0 | 0 | — |
case-05 | pass→pass | 13,922 | 10,142 | -27% | 1 | 1 | 0% | 2,279 | 2,465 | +8% | 0 | 0 | — |
case-06 | pass→pass | 13,230 | 10,796 | -18% | 1 | 1 | 0% | 2,494 | 2,695 | +8% | 0 | 0 | — |
case-07 | pass→pass | 4,395 | 4,458 | +1% | 1 | 1 | 0% | 735 | 1,579 | +115% | 0 | 0 | — |
case-08 | pass→pass | 10,040 | 4,892 | -51% | 1 | 1 | 0% | 1,864 | 1,684 | -10% | 0 | 0 | — |
case-09 | pass→pass | 9,957 | 10,932 | +10% | 1 | 1 | 0% | 1,604 | 2,769 | +73% | 0 | 0 | — |
case-10 | fail→pass | 5,965 | 3,764 | -37% | 1 | 1 | 0% | 1,090 | 1,448 | +33% | 0 | 0 | — |
case-11 | fail→pass | 6,013 | 4,005 | -33% | 1 | 1 | 0% | 934 | 1,492 | +60% | 0 | 0 | — |
case-12 | pass→pass | 14,170 | 9,638 | -32% | 1 | 1 | 0% | 2,474 | 2,498 | +1% | 0 | 0 | — |
case-13 | pass→pass | 6,014 | 2,763 | -54% | 1 | 1 | 0% | 1,073 | 1,229 | +15% | 0 | 0 | — |
case-14 | pass→pass | 16,251 | 14,972 | -8% | 1 | 1 | 0% | 2,642 | 3,205 | +21% | 0 | 0 | — |
case-15 | fail→pass | 11,639 | 4,728 | -59% | 1 | 1 | 0% | 2,185 | 1,675 | -23% | 0 | 0 | — |
case-16 | pass→pass | 11,585 | 11,860 | +2% | 1 | 1 | 0% | 2,191 | 3,041 | +39% | 0 | 0 | — |
case-17 | pass→pass | 16,649 | 8,403 | -50% | 1 | 1 | 0% | 1,697 | 2,040 | +20% | 0 | 0 | — |
case-18 | pass→pass | 5,058 | 3,240 | -36% | 1 | 1 | 0% | 750 | 1,244 | +66% | 0 | 0 | — |
case-19 | pass→pass | 5,161 | 2,187 | -58% | 1 | 1 | 0% | 806 | 1,169 | +45% | 0 | 0 | — |
case-20 | fail→fail | 15,977 | 21,816 | +37% | 1 | 1 | 0% | 2,768 | 4,017 | +45% | 0 | 0 | — |
case-21 | fail→fail | 17,609 | 11,998 | -32% | 1 | 1 | 0% | 3,003 | 2,937 | -2% | 0 | 0 | — |
case-22 | pass→pass | 16,365 | 9,357 | -43% | 1 | 1 | 0% | 2,847 | 2,462 | -14% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +18 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.