Security smell detection on Combined (Puppet, Ansible, Chef) oracle (test)
83.1Macro F1IntelliSA
Evaluation Results
| Method | Links | |
|---|---|---|
| IntelliSA2026.01 | 83.1 | |
| Grok-4model=Grok-4-fast, Prompting style=static-analysis-rules2026.01 | 75.7 | |
| GLITCH2026.01 | 69.4 | |
| Grok-4model=Grok-4-fast, Prompting style=definition-based2026.01 | 68.7 | |
| GPT-5model=GPT-5-2025-08-07, Prompting style=static-analysis-rules2026.01 | 65.3 | |
| Claude-4model=Claude-Sonnet-4.0, Prompting style=definition-based2026.01 | 62.2 | |
| Claude-4model=Claude-Sonnet-4.0, Prompting style=static-analysis-rules2026.01 | 55.9 | |
| GPT-5model=GPT-5-2025-08-07, Prompting style=definition-based2026.01 | 52.7 | |
| SLAC/SLIC2026.01 | 41.3 |