openai/gpt-6-luna
Each row gives the model more help than the one above it.
| Context | Runs | Design system | Quality | Done | Cost per run |
|---|---|---|---|---|---|
| none file tools only | 7 | 141.7 | 26.7 | 86% | $0.0034 |
| storybook-mcp Storybook's own MCP | 5 | 36.2 | 18 | 100% | $0.0047 |
| ds-mcp our design-system MCP | 7 | 2.1 | 16.1 | 86% | $0.0028 |
| ds-mcp-check the MCP and a checker | 5 | 0 | 20 | 100% | $0.0055 |
Counted violations over all its runs.
| Rule | Group | Violations | Runs |
|---|---|---|---|
| raw-values/raw-length | design-system | 701 | 17 |
| raw-values/raw-color | design-system | 428 | 10 |
| lint/@pandacss(no-physical-properties) | quality | 296 | 21 |
| typecheck/TS2769 |
| ds-mcp-check-full a checker with compiler and linter | 5 | 0 | 0 | 100% | $0.0132 |
| quality |
| 124 |
| 18 |
| raw-values/raw-number | design-system | 23 | 3 |
| a11y/color-contrast | quality | 19 | 10 |
| components/native-element | design-system | 19 | 6 |
| token-roles/palette | design-system | 14 | 1 |
| typecheck/TS18048 | quality | 11 | 4 |
| lint/typescript(no-unsafe-type-assertion) | quality | 8 | 6 |