mahmutademm/investment-analyst
Top-down equity research skills: macro cycle and sector rotation (macro-analysis), sub-industry screening (sector-screening), 10-K deep dives with earnings-quality, credit and valuation workbooks (pick-the-stock), and position tracking (portfolio-review), backed by a Python toolkit (FRED, Yahoo Finance, SEC EDGAR) and Excel. The user makes every decision. Start with investment-setup.
First-run setup and maintenance for the investment-analyst skills (Claude Code, Codex, Cursor, Copilot, Gemini CLI and other Agent Skills tools). Installs the Python toolkit, creates the user's private investment workspace, records the SEC contact string and portfolio rules, runs a health check and smoke test, and optionally connects a private GitHub repo with scheduled Actions. Use when the user asks to set up the investment analyst, says "set up", "first run", "install", "update the plugin", when any other investment-analyst skill reports "No investment workspace found", or when the user wants to change sectors, thresholds or schedules. Takes an optional mode (first-run, update, github or doctor).
Top-down macro and sector-rotation analysis for a ~6-month horizon. Reads inflation (CPI, core PCE), the Fed path (actual and market-implied), the 10-year yield and curve, credit spreads, labor and activity data; proposes a business-cycle phase; scores every sector on four pillars (cycle fit, earnings revisions, relative strength, valuation vs its own history); and helps the user set sector weights and the sub-industries to screen. Produces Macro_<date>.xlsx and records the decisions for the screening skill. Use for "what cycle are we in", "which sectors should I be in", "sector rotation", "monthly scorecard review". Takes an optional mode (full, review or scorecard-only).
Deep dive on 1-4 stocks the user chose. Pulls SEC XBRL financials and the filing pack (10-K, 10-Q, proxy, earnings releases, 8-Ks), reads the filings, builds business/moat and a SWOT tied to model assumptions, runs the earnings-quality and credit gate (5-step DuPont, CFO/NI, accruals, Beneish, Piotroski, Altman, leverage, coverage, maturities), compares relative value with the full sub-industry peer median and the stock's own history, values it by life-cycle stage (3-stage DCF, reverse DCF, scenarios and sensitivity, DDM, residual income, normalized earnings), and prepares a decision sheet (6-month expected return, catalyst, ATR stop, position size, scale-in). Produces <TICKER>_DeepDive_<date>.xlsx. Takes 1-4 tickers. Use for "deep dive", "analyze the 10-K", "is X a buy", "value X".
Ongoing tracking for the investment-analyst workflow. Reads the daily/weekly reports from GitHub Actions (or runs them), checks every watchlist name against its price stop, thesis stop, sector relative strength, catalyst and earnings dates, flags scale-in tranche triggers, runs the monthly scorecard re-check, and records decisions in the journal. Use for "check my positions", "did anything trigger", "weekly review", "monthly review", "update my watchlist". Takes an optional mode (daily, weekly or monthly).
Mechanical, peer-relative screening of the sub-industries chosen in macro-analysis (valuation, quality, estimate revisions, momentum, balance sheet, forensic flags), followed by a sourced "why is it cheap?" news check on the top names (temporary vs structural, catalyst inside 6 months). Produces Screen_<key>_<date>.xlsx. The user picks the 3-4 stocks for the deep dive. Use for "screen <sub-industry>", "find candidates in <sector>", "which stocks in medical devices". Takes optional sub-industry keys.