独立主控底盘型成长股正式研究；用私有 modules、多角色/subagent、中间产物、进攻型市场重定价主线、行业经济性与蛋糕分配、竞争/客户验证链、中期进攻 QA、终审事实 QA 和投资人写作，完整覆盖旧业务底盘、第二成长曲线、TAM/SAM/SOM、平台复用、承接动作、利润桥、跟踪体系、证据边界、三表/FCF 建模接力输入和估值接力输入。Use when 用户需要深度分析 A 股制造业、AI算力、数据中心、新能源、机器人、先进制造、半导体设备/材料等底盘型成长公司，判断旧业务托底、行业经济性拐点、新业务抬天花板、公司份额提升、平台能力复用、竞争客户验证、利润中枢上移、预期差和证伪路径，并产出正式深度报告而非摘要。
独立主控供应链平台型成长股正式研究；用私有 modules、多角色/subagent、中间产物、进攻型市场交易变量、技术路线到价值量、原材料价格链、竞争/客户认证链、产能升级、订单经营验证、利润天花板、跟踪体系、两轮 QA 和投资人写作，完整覆盖 AI 算力、数据中心、新能源、机器人、先进封装、半导体设备/材料等供应链公司研究。Use when 用户需要深度分析供应链公司，判断身份切换、订单放量、技术路线升级、利润中枢上移、天花板抬升、市场预期差和证伪路径，并产出正式深度报告而非摘要。
Real DCF (Discounted Cash Flow) model creation for equity valuation. Retrieves financial data from SEC filings and analyst reports, builds comprehensive cash flow projections with proper WACC calculations, performs sensitivity analysis, and outputs professional Excel models with executive summaries. Use when users need to value a company using DCF methodology, request intrinsic value analysis, or ask for detailed financial modeling with growth projections and terminal value calculations.
DCF-only valuation workflow orchestrator for equity research. Use when the user asks for DCF估值, DCF闭环, intrinsic value, Formal DCF, Scenario DCF, Reverse DCF, or wants a one-pass workflow that first prepares financial-modeling DCF-ready inputs and then runs dcf-model outputs. This skill coordinates financial-modeling to dcf-model for DCF deliverables only.
Build integrated financial models with 3-statement projections, DCF-ready UFCF bridges, working-capital schedules, and debt/interest linkages. Use for income statement, balance sheet, cash flow, and DCF input preparation. Scope is three-statement and DCF-ready modeling; PEG-ready packages are handled by growth-stock-valuation.
成长股 PEG/动态 PE 估值 skill；用于成长股、供应链平台股、产业链龙头的 PEG、动态 PE、复合 PEG、目标市值、目标价、一致预期差、当前市值隐含预期、估值年份切换和证伪点分析。目标市值默认使用自有正常/乐观利润锚，一致预期只作市场对照；范围限定为 PEG/动态 PE 成长定价。若用户要求 bottom-up/算小账/分业务利润桥与 PEG 并列对照，使用共享 catalyst precheck、独立 PEG subagent、独立 bottom-up subagent 和并列模块，防止估值与利润计算互相污染。
Sync local Codex skills from ~/.codex/skills into a GitHub skills repository, update README, commit, and optionally push.
Build, migrate, inspect, and operate the local Feishu-to-Codex research bridge: Feishu bot messages create queued research tasks, a serialized worker runs Codex with research skills, final reports are converted to PDF or Drive links, and results are sent back to a Feishu group. Use when the user mentions 飞书/Codex 桥梁, 飞书投研队列, lark-cli event consume, /投研, queue worker, report delivery to Feishu, or migrating this bridge to another repository.