Stock Analyzer: Agentic Workflow Orchestration

A research-first stock analysis toolkit designed for AI agents, built around three institutional-grade .agents workflows.

This repository combines workflow orchestration, analyst-style financial modeling (DCF and Comps), and structured markdown generation to support single-ticker initiation research, portfolio reviews, and secular trend analysis.

Agentic Execution & Core Workflows

Agents can trigger these primary workflows via their corresponding slash commands. All heavy lifting is handled by centralized Python scripts located in .agents/skills/.

1. /ticker-analysis

Responsible for deep, multi-skill initiation research on a single ticker.

• What it does:
  • Gathers fundamental, momentum, and risk metrics using compute_indicators.py.
  • Runs DCF and relative valuation (Comps) via Python automation.
  • Checks for "Relative Oversold" institutional dip setups using find_dip.py.
  • Produces an institutional-quality out/[TICKER]_Initiation_Report.md.
• When to use:
  • You need a full research note for one specific stock.
  • You want definitive actionable guidance for new buyers vs. current holders.

2. /short-term-portfolio-review

Designed to batch-review a portfolio of tickers and create a tactical short-term action report (1-3 month horizon).

• What it does:
  • Reads the portfolio list from ticker-list-s.txt.
  • Extracts the exact CAD exchange pricing using fetch_cad_price.py to prevent currency hallucination.
  • Uses invoke_subagent to run /ticker-analysis for each position in parallel.
  • Synthesizes a portfolio-level strategy report in out_master/[YYYY-MM-DD]/Short_Term_Portfolio_Review.md based on absolute position sizing and technical confluence.
• When to use:
  • You need a tactical review of a live portfolio to determine precise Add on Dip, Hold, or Trim targets.

3. /secular-analysis

Built for long-term, macro-driven secular trend investing.

• What it does:
  • Analyzes sector tailwinds and the company’s value chain position.
  • Screens for moat strength, capital intensity, and structural risk.
  • Builds a falsifiable thesis with kill criteria and DCF support.
  • Produces out_secular/[TICKER]_Secular_Datapack.md and out_secular/[TICKER]_Secular_Playbook.md.
• When to use:
  • You are evaluating a stock as a long-term core holding (1+ year timeframe).

Repository Structure & Stack

• .agents/workflows/ — Workflow definitions (YAML/Markdown) for each agent pipeline.
• .agents/skills/ — Reusable skill modules (Python scripts) powering the workflows.
  • Tech Stack: Python, uv, yfinance, pandas, pandas_ta
• out/ — Single-ticker outputs and intermediate deliverables (Excel & Markdown).
• out_master/ — Master portfolio review strategy documents.
• out_secular/ — Long-term secular analysis deliverables.

Inputs & Configuration

• ticker-list-s.txt: The core input file for the /short-term-portfolio-review workflow. It must follow a structured CSV format so agents can parse it accurately: Ticker, Name, Avg(CAD), Shares (e.g., IBM.TO, International Business Machines Corporation, 40.9, 26)