Command-line Operation Execution on GitHub-CLI
94.1Success RateANYTHING2SKILL + RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| ANYTHING2SKILL + RAGBase Model=gpt5.4, Context=SkillBank + RAG2026.06 | 94.1 | |
| ANYTHING2SKILLBase Model=gpt5.4, Context=SkillBank2026.06 | 82.3 | |
| RAGBase Model=gpt5.4, Context=top-3 source entries2026.06 | 76.5 | |
| Base AgentBase Model=gpt5.4, Context=None2026.06 | 64.7 |