> For the complete documentation index, see [llms.txt](https://nfa.gitbook.io/nfa-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://nfa.gitbook.io/nfa-docs/introduction.md). # Introduction

## NFA · Forecasting engine for prediction markets NFA turns a prediction market into an independent, auditable probability. You select a market from the curated feed of markets NFA is built to forecast (today, on Polymarket). It researches the question into an evidence file, casts one AI advocate for each outcome, runs a structured debate, and an impartial judge weighs the cases against the evidence. The number it returns is a genuine second opinion rather than an echo of the crowd. {% hint style="info" %} **NFA is a tool, not an oracle.** It hands you a probability and a full reasoning trace, then gets out of the way. It never places a trade for you and never tells you what to do. The call, and the risk, are always yours. Not Financial Advice. {% endhint %} ### What NFA is Most traders price a market on instinct, or they read the crowd price and anchor to it. NFA does neither. It builds an evidence file from live sources, argues every side in the open, and grades the debate into a calibrated distribution you can audit line by line. Then it checks its own work: a separate step can veto the forecast, and every resolved market is scored against the market's own price, so the track record is honest rather than flattering. ### The shape of the system Four layers sit around one core. ```mermaid flowchart TD U[Trader or AI agent] --> B[Binding] M[Prediction market] --> B B --> D[Advocate debate] A[Live data anchors] --> D D --> J[Judge] J --> F[Forecast and reasoning trace] F --> V[Self-verification] V --> Q[Quality badge] F --> L[Accuracy record on resolution] F --> T[Optional non-custodial trade] ``` * **The engine** researches, debates, and judges a market into a probability distribution. * **The data layer** anchors forecasts to real measurements, from central-bank rates to open-source conflict maps. * **The accuracy layer** verifies each forecast and scores it against the market after resolution. * **The surfaces** put the engine in reach: a web app, with an MCP server for AI agents on the way. ### Explore the reference


The forecasting engine	How a market becomes a calibrated probability.
Live data and adapters	The data that anchors forecasts to reality.
The market lifecycle	Discovery, binding, resolution, and honest scoring.
Trading and portfolio	Act on a forecast, non-custodially.
The economy	Credits, the $NFA token, and rewards.
Interfaces	The web app and how agents will connect.

New here? Start with Concepts and glossary, then read The forecast pipeline. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://nfa.gitbook.io/nfa-docs/introduction.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.