Everything from chapters 1 through 8 collapses into one end-to-end workflow. Three companies, one decision. This chapter IS the workshop — expected time: 4 to 8 hours across multiple sittings.

A single agent is a tool. A fleet with defined roles, cross-verification, and explicit handoffs is transformative. Five patterns cover 90% of investment workflows — but over-engineering the fleet is itself a form of procrastination.

Most investors have beliefs, not theses. A written thesis with kill criteria is accountability against your own memory — because memory is self-serving, and beliefs mutate silently.

Risk is the probability of a loss you cannot afford — not standard deviation. LTCM optimized for volatility and was killed by correlation breakdown. Four real risks, and why only you can decide what you can afford to lose.

Every number in a DCF is a story choice. Agents compute flawlessly and default to consensus. Your job is the narrative — and knowing which assumption your valuation secretly depends on.

You do not need to read every line of a 10-K. You need to recognize patterns. Three financial statements as three lenses — business structure, fragility, and truth.

A stock is a claim on future cash flows — not a ticker, not a chart. Three sources of return, five types of edge, and the vocabulary you need before you can judge an agent's output.

Most people use their AI agent in one mode. Investment agents play four distinct roles — analyst, red team, tutor, executor — and the way you frame your request determines which role you get.

Investment analysis used to require either a Bloomberg terminal or the ability to write code. AI agents removed both prerequisites. The barrier is now judgment — and that changes what an investing curriculum should teach.

I built a personal knowledge compiler that treats the LLM as a programmer and your wiki as a codebase. 15 source files, 5 dependencies, zero databases.