references/research-methodology.md | DIKW 模型、假设驱动法、认知偏差清单、信息源可信度标准 | industry-research, stock-fundamental |
references/external-factors.md | PEST 框架、政策信号分类、经济周期传导、Gartner 技术曲线 | industry-research |
references/lifecycle-framework.md | 渗透率标尺、四阶段框架(导入/成长/成熟/衰退)、关键阈值、阶段与分析工具对应表 | industry-research, stock-fundamental |
references/data-quality-levels.md | P1-P5 分级与使用规则、数据冲突处理、AI 数据幻觉提醒 | industry-research, stock-fundamental |
references/market-sizing-principles.md | 双重验证原则、三种测算方法、无公开数据另类路径、常见测算错误 | industry-research |
references/moat-framework.md | 护城河两大类型、强弱信号、假护城河清单、持续性判断、与生命周期关系 | industry-research, stock-fundamental |
references/pitfalls.md | 十二大通用陷阱、专家访谈三步处理法 | industry-research, stock-fundamental |
references/time-consistency.md | 时间口径匹配规则、输出标注要求、常见违规情形 | industry-research, stock-fundamental |
references/scqr-principle.md | SCQR 汇报顺序、受众对应结构、结论质量要求、输出深度分级 | industry-research, stock-fundamental |
references/research-discipline-and-debiasing.md | 研究纪律最高纲领、四段式工作流、去偏清单、关键假设表、grounding 纪律、actionable 禁令 | stock-fundamental, theme-chain-context, event-interpretation |
references/cognitive-edge-framework.md | 信息差/逻辑差/预期差/催化剂四层博弈差、完整性合成判断、禁制造虚假博弈差 | stock-fundamental, theme-chain-context |
references/profit-attribution.md | rent 7 类截留者、rent capture % 判定、持续性、产业利好 vs 公司利好 | stock-fundamental, theme-chain-context |
references/source-tier-and-snapshot-rules.md | A股来源→P1-P5 映射、按类型匹配级别、引用格式、快照数据防过时规则、口径标注 | industry-research, stock-fundamental, theme-chain-context, event-interpretation |
references/deep-research-card.md | 深研卡片统一输出契约:人读结构 + canonical 机读 schema + 枚举取值,汇总各模块机读块供下游回填 | stock-fundamental, theme-chain-context, event-interpretation |
references/valuation-paradigm.md | A股按资产类型选估值法、A股独有维度(历史分位/股息/政策/题材市梦率/解禁减持/AH溢价)、估值输出边界(禁目标价) | stock-fundamental, industry-research |
references/downstream-orchestration.md | 深研层编排契约:调用顺序、C1 输入、C8 输出、各模块职责、行业复用 | 下游编排方/开发者参考 |