What problem does it solve? Producing a reproducible, source-cited investment research report for a listed company requires coordinating many specialized analyses (financials, valuation, technicals, macro, risk) plus QA, which is error-prone when done ad hoc. This Skill normalizes user input, fans work out to expert analyst roles, reconciles conflicting numeric claims, and assembles a QA-reviewed final report. ## Core Features & Use Cases - Input Intake Gate: Auto-identifies a company from just a name or ticker, confirms exchange, currency, and accounting basis, and asks the user only five option questions with sensible defaults. - Fan-out/Fan-in Pipeline: Dispatches six analyst roles (financial, fundamental, valuation, technical, macro-sentiment, risk-scenario) plus section writers, then synthesizes an 18-section draft report. - Stage Gates & Reconciliation: Enforces S0–S8 stage gates with machine-checkable entry/exit criteria, quarantines divergent numeric claims, and supports targeted re-runs driven by the QA fix-list. - Use Case: Ask for a deep-dive report on a ticker like CSTM with a tariff catalyst focus; the Skill produces a workspace of evidence ledgers, per-section findings, a QA-reviewed final Markdown report, and an interactive HTML dashboard. ## Quick Start Ask the assistant to generate a full investment research report for a company by giving its name or ticker, for example requesting a deep analysis of AAPL with default options.
