學術辯論與觀點比較技能

純指令技能。LLM 結構化思考 + read_draft, search_local_references, mcp_cgu_deep_think, mcp_cgu_spark_collision。

觸發：辯論、pro/con、devil's advocate、挑戰假設、compare viewpoints

Framework 1: Academic Debate（雙方辯論）

針對研究議題結構化正反方論點。按證據等級排列： SR/Meta > RCT > Cohort > Case-Control > Cross-Sectional > Case Series > Expert Opinion

輸出：Position A (supporting) → Position B (opposing) → Methodological Considerations (偏誤) → Synthesis (agreement / disagreement / clinical bottom line)

系統性挑戰一個研究主張。5 面向：

Methodological — 研究類型對應偏誤（見下方）
Statistical — multiple comparisons, effect size, CI width, missing data, power
Generalizability — population, setting, timeframe, intervention fidelity
Alternative Explanations — confounding, reverse causation, temporal trends, Hawthorne

純指令技能。LLM 結構化思考 + read_draft, search_local_references, mcp_cgu_deep_think, mcp_cgu_spark_collision。

觸發：辯論、pro/con、devil's advocate、挑戰假設、compare viewpoints

針對研究議題結構化正反方論點。按證據等級排列： SR/Meta > RCT > Cohort > Case-Control > Cross-Sectional > Case Series > Expert Opinion

輸出：Position A (supporting) → Position B (opposing) → Methodological Considerations (偏誤) → Synthesis (agreement / disagreement / clinical bottom line)

系統性挑戰一個研究主張。5 面向：

Methodological — 研究類型對應偏誤（見下方）
Statistical — multiple comparisons, effect size, CI width, missing data, power
Generalizability — population, setting, timeframe, intervention fidelity
Alternative Explanations — confounding, reverse causation, temporal trends, Hawthorne

Type	Key Biases
RCT	Selection, performance, detection, attrition, reporting
Cohort	Selection, confounding, information, loss to follow-up, healthy worker
Case-Control	Recall, selection (controls), confounding, misclassification, temporal ambiguity
Cross-Sectional	No temporality, prevalence-incidence, non-response, information
Retrospective	Information (records), survivorship, confounding, missing data