Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Check if code is readable by non-developers - clear names, plain English comments, no jargon
.claude/skills/aiskillstore-code-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -63% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -62% | 0% |
Analyzes code to ensure non-developers (managers, stakeholders, new team members) can understand it.
bashpython3 analyze.py --path your-file.py --strictness lenient
Bad Code (score: 71/100):
pythondef proc(usr, tkn): tmp = usr + tkn return tmp * 2
Issues: Cryptic names, no comments, unclear purpose.
Good Code (score: 95/100):
pythondef process_user_authentication(username, auth_token): """Validate user credentials and return auth score""" combined_credential = username + auth_token return combined_credential * 2
--strictness lenient to reduce noise| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 7,564 | 3,237 | -57% | 1 | 1 | 0% | 1,086 | 821 | -24% | 0 | 0 | — |
case-02 | fail→fail | 6,510 | 5,049 | -22% | 1 | 1 | 0% | 1,073 | 1,004 | -6% | 0 | 0 | — |
case-03 | fail→fail | 2,018 | 2,692 | +33% | 1 | 1 | 0% | 321 | 696 | +117% | 0 | 0 | — |
case-04 | fail→fail | 7,191 | 11,543 | +61% | 1 | 1 | 0% | 939 | 1,831 | +95% | 0 | 0 | — |
case-05 | pass→pass | 12,282 | 10,912 | -11% | 1 | 1 | 0% | 2,149 | 2,137 | -1% | 0 | 0 | — |
case-06 | pass→pass | 19,571 | 23,284 | +19% | 1 | 1 | 0% | 4,294 | 5,478 | +28% | 0 | 0 | — |
case-07 | fail→fail | 13,796 | 12,600 | -9% | 1 | 1 | 0% | 2,078 | 2,171 | +4% | 0 | 0 | — |
case-08 | pass→pass | 12,211 | 11,268 | -8% | 1 | 1 | 0% | 1,899 | 2,089 | +10% | 0 | 0 | — |
case-09 | fail→pass | 9,596 | 2,040 | -79% | 1 | 1 | 0% | 1,521 | 562 | -63% | 0 | 0 | — |
case-10 | pass→pass | 14,234 | 6,257 | -56% | 1 | 1 | 0% | 2,189 | 1,260 | -42% | 0 | 0 | — |
case-11 | pass→pass | 11,058 | 2,695 | -76% | 1 | 1 | 0% | 1,669 | 628 | -62% | 0 | 0 | — |
case-12 | fail→pass | 8,780 | 2,025 | -77% | 1 | 1 | 0% | 1,355 | 547 | -60% | 0 | 0 | — |
case-13 | fail→pass | 13,770 | 12,620 | -8% | 1 | 1 | 0% | 1,995 | 2,177 | +9% | 0 | 0 | — |
case-14 | fail→pass | 5,124 | 2,038 | -60% | 1 | 1 | 0% | 801 | 596 | -26% | 0 | 0 | — |
case-15 | fail→pass | 9,504 | 2,044 | -78% | 1 | 1 | 0% | 1,458 | 554 | -62% | 0 | 0 | — |
case-16 | pass→pass | 8,404 | 4,011 | -52% | 1 | 1 | 0% | 1,335 | 915 | -31% | 0 | 0 | — |
case-17 | pass→pass | 11,914 | 10,584 | -11% | 1 | 1 | 0% | 1,804 | 1,974 | +9% | 0 | 0 | — |
case-18 | fail→pass | 8,518 | 2,591 | -70% | 1 | 1 | 0% | 1,410 | 669 | -53% | 0 | 0 | — |
case-19 | fail→pass | 7,837 | 2,716 | -65% | 1 | 1 | 0% | 1,222 | 731 | -40% | 0 | 0 | — |
case-20 | pass→pass | 11,581 | 6,091 | -47% | 1 | 1 | 0% | 1,659 | 1,257 | -24% | 0 | 0 | — |
case-21 | pass→pass | 7,658 | 5,322 | -31% | 1 | 1 | 0% | 1,223 | 1,067 | -13% | 0 | 0 | — |
case-22 | pass→pass | 10,946 | 7,023 | -36% | 1 | 1 | 0% | 1,755 | 1,422 | -19% | 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 +32 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.