what is equity-research-skill fr?

This project is an AI agent skill, meaning a set of instructions and reference files that a tool like Claude Code can load, for producing deep equity research reports on individual stocks. Rather than a short summary, it is built to make an AI follow a discipline closer to what a professional investment research team would use: every fact should be traceable to a source, every valuation should be able to be recalculated, and every conclusion should be checked against the numbers behind it. The skill has two main modes. A full research mode produces a nine section report covering the company's business, competitive position, management, financials, valuation, analyst views, catalysts and risks, and an investment conclusion. An earnings mode is meant to be used right after a company reports quarterly or annual results, and produces its own nine section breakdown covering how the results compare to expectations, revenue and segment detail, margins, cash flow, management's guidance and call commentary, and how the valuation model changes as a result. If there is no existing research on the company yet, earnings mode will first rebuild a multi year history before analyzing the new results. For valuation specifically, the skill requires cross checking with at least three separate methods, such as discounted cash flow, reverse discounted cash flow, scenario weighted valuation, and earnings power value, and it runs the actual calculations through an included Python script rather than asking the AI to estimate a number. It also ships eight industry specific appendices, for sectors like SaaS, semiconductors, banks, insurance, pharmaceuticals, consumer goods, energy, and utilities, each changing which metrics and valuation approach get used. To install it, a user copies the repository into a folder that their AI tool reads skills from, such as Claude Code's skills directory, or simply points any agent capable of reading files at the SKILL.md file. Running the valuation script needs some Python environment, though not necessarily one installed on the user's own machine, since it only relies on the standard library. The skill states clearly that its output is for research reference only, not investment advice, and it is released under the MIT license.