Install any skill in seconds. Free to start, no credit card required.
Get Started Free →MISRA C compliance checking and static analysis integration
.claude/skills/a5c-ai-misra-c-analysis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -50% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -23% | 0% |
This skill provides MISRA C compliance checking and static analysis integration for safety-critical embedded software development, ensuring code quality and standard compliance.
misra-c-compliance.js - MISRA C compliance checkingfunctional-safety-certification.js - Safety standard compliancedevice-driver-development.js - Driver code qualityThis skill is invoked when tasks require:
Must always be followed; no deviations permitted.
Must be followed unless formally deviated with documented justification.
Should be followed; deviations need documentation but less formal.
Guidelines that cannot be fully verified by static analysis.
// MISRA C:2012 configuration
+e9*, +e1*, +e2*, +e3*
-append(9001,[MISRA 2012 Rule 1.1, required])
-append(9002,[MISRA 2012 Rule 1.2, advisory])xml<misra> <rule number="1.1" severity="error"/> <rule number="1.2" severity="warning"/> <suppress rule="15.5" file="legacy/*.c"/> </misra>
markdown## Deviation Record DR-001 **Rule**: MISRA C:2012 Rule X.Y **Location**: file.c, line 123 **Justification**: [Technical reason for deviation] **Risk Assessment**: [Impact and mitigation] **Approval**: [Approver name and date]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,188 | 23,121 | +15% | 1 | 1 | 0% | 3,764 | 4,416 | +17% | 0 | 0 | — |
case-02 | pass→pass | 31,069 | 41,962 | +35% | 1 | 1 | 0% | 5,471 | 6,934 | +27% | 0 | 0 | — |
case-03 | pass→pass | 18,880 | 16,610 | -12% | 1 | 1 | 0% | 2,097 | 3,455 | +65% | 0 | 0 | — |
case-04 | fail→fail | 15,529 | 13,811 | -11% | 1 | 1 | 0% | 2,374 | 3,313 | +40% | 0 | 0 | — |
case-05 | pass→pass | 13,087 | 15,023 | +15% | 1 | 1 | 0% | 2,304 | 2,859 | +24% | 0 | 0 | — |
case-06 | pass→pass | 12,193 | 13,155 | +8% | 1 | 1 | 0% | 2,065 | 2,854 | +38% | 0 | 0 | — |
case-07 | pass→pass | 16,218 | 18,991 | +17% | 1 | 1 | 0% | 2,792 | 3,487 | +25% | 0 | 0 | — |
case-08 | fail→pass | 10,117 | 5,290 | -48% | 1 | 1 | 0% | 1,977 | 1,622 | -18% | 0 | 0 | — |
case-09 | fail→pass | 12,319 | 1,913 | -84% | 1 | 1 | 0% | 2,021 | 1,002 | -50% | 0 | 0 | — |
case-10 | fail→pass | 10,161 | 2,620 | -74% | 1 | 1 | 0% | 1,670 | 957 | -43% | 0 | 0 | — |
case-11 | fail→pass | 12,985 | 1,848 | -86% | 1 | 1 | 0% | 1,978 | 881 | -55% | 0 | 0 | — |
case-12 | fail→pass | 17,165 | 6,110 | -64% | 1 | 1 | 0% | 2,361 | 1,810 | -23% | 0 | 0 | — |
case-13 | pass→pass | 14,642 | 8,586 | -41% | 1 | 1 | 0% | 2,152 | 2,253 | +5% | 0 | 0 | — |
case-14 | pass→pass | 21,600 | 15,906 | -26% | 1 | 1 | 0% | 3,189 | 3,051 | -4% | 0 | 0 | — |
case-15 | fail→pass | 17,866 | 3,191 | -82% | 1 | 1 | 0% | 3,188 | 1,156 | -64% | 0 | 0 | — |
case-16 | fail→pass | 12,058 | 7,720 | -36% | 1 | 1 | 0% | 1,837 | 2,216 | +21% | 0 | 0 | — |
case-17 | pass→pass | 25,267 | 4,627 | -82% | 1 | 1 | 0% | 2,945 | 1,644 | -44% | 0 | 0 | — |
case-18 | pass→pass | 15,438 | 16,734 | +8% | 1 | 1 | 0% | 2,233 | 3,080 | +38% | 0 | 0 | — |
case-19 | pass→pass | 15,782 | 18,814 | +19% | 1 | 1 | 0% | 2,488 | 3,437 | +38% | 0 | 0 | — |
case-20 | pass→pass | 21,245 | 22,888 | +8% | 1 | 1 | 0% | 3,541 | 3,984 | +13% | 0 | 0 | — |
case-21 | pass→pass | 16,729 | 24,770 | +48% | 1 | 1 | 0% | 2,532 | 4,387 | +73% | 0 | 0 | — |
case-22 | pass→fail | 20,074 | 20,493 | +2% | 1 | 1 | 0% | 3,472 | 4,397 | +27% | 0 | 0 | — |
case-23 | pass→pass | 13,272 | 4,898 | -63% | 1 | 1 | 0% | 2,358 | 1,536 | -35% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.