원클릭으로
model-only
Route non-tool questions to the language model with explicit no-tool labeling.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Route non-tool questions to the language model with explicit no-tool labeling.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
Analyze mutations by comparing wild-type and mutant sequences. Use this skill when: (1) User provides WT and MT sequences, (2) User asks about mutation impact (e.g., "p.R175H"), (3) User wants to compare two sequences, (4) User asks "what's the difference between these sequences".
Search PubMed for biomedical research papers by topic, disease, or gene. Use this skill when: (1) User asks to find papers about a topic, (2) User wants recent literature on a disease or gene, (3) User needs citations for research, (4) User asks "papers about X" or "literature on Y".
Analyze DNA/RNA/protein sequences and optionally probe UniProt for homology. Use when user provides FASTA format or raw sequence.
Query UniProt database for protein information including function, organism, sequence, and structure. Use this skill when: (1) User provides a UniProt accession (e.g., P38398, Q9H3P7), (2) User asks about a protein by name (e.g., "BRCA1", "hemoglobin"), (3) User wants protein function, organism, or sequence information, (4) User needs to find similar proteins by sequence search.
SOC 직업 분류 기준
| name | model_only |
| description | Route non-tool questions to the language model with explicit no-tool labeling. |
| category | meta |
| tools_used | [] |
Use this skill only after first-level routing determines that the user is not asking for a concrete scientific lookup or computation.
Appropriate questions include:
Do not use this skill for concrete biological facts that should be grounded in tools, including gene/protein lookup, sequence analysis, mutation assessment, literature search, database queries, similarity search, pathways, structures, clinical evidence, or compound/target lookup.
Outputs must state that no external scientific tools or database evidence were used. If the user actually needs database-backed evidence, route to the relevant scientific skill instead.
Hybrid fallback rule:
0.0, Open-Rosalind may append a clearly labeled model-only
fallback.