ワンクリックで
theoretical-physics-agents
theoretical-physics-agents には RintaroMasaoka から収集した 15 個の skills があり、リポジトリ単位の職業カバレッジとサイト内 skill 詳細ページを表示します。
このリポジトリの skills
Improve agent prompts and behavior. The user says what's unsatisfactory; AI understands the complaint deeply, rewrites at the root cause, and an independent reviewer verifies.
Improve agent prompts and behavior. The user says what's unsatisfactory; AI understands the complaint deeply, rewrites at the root cause, and an independent reviewer verifies.
Autonomously execute research cycles. Specify the cycle limit as an argument (e.g., /auto 2). Default: 5.
Review research direction and AI reliability with the user, record oversight decisions, and update the next research focus.
Execute the normal /auto research loop with a human approval checkpoint after each planner update and before execution.
Autonomously execute research cycles. Specify the cycle limit as an argument (e.g., /auto 2). Default: 5.
Review research direction and AI reliability with the user, record oversight decisions, and update the next research focus.
Execute the normal /auto research loop with a human approval checkpoint after each planner update and before execution.
Synchronize framework-level changes between a child research project and the upstream theoretical-physics-agents framework. Use for pulling framework files before editing them or pushing prompt/script/config/workflow improvements upstream.
Autonomously execute paper writing cycles. Specify the cycle limit as an argument (e.g., /write 3). Default: 5.
Autonomously execute paper writing cycles. Specify the cycle limit as an argument (e.g., /write 3). Default: 5.
Set or change the research theme and direction.
Set or change the research theme and direction.
Fetch arXiv papers (PDF + LaTeX source) via GitHub Actions relay. Usage: /fetch-arxiv 2301.00001 [cond-mat/9905027 ...]
Fetch arXiv papers (PDF + LaTeX source) via GitHub Actions relay. Usage: /fetch-arxiv 2301.00001 [cond-mat/9905027 ...]