AI Trader

What this project is

AI Trader is a browser-first commodities operator desk built for advisory research, signal review, paper-trade proposals, and AI-assisted market context rather than real-money execution. It is deliberately focused on commodities workflows instead of generic market-dashboard sprawl, with charts, signals, risk framing, research surfaces, macro and prediction-market context, and an AI Desk layer intended to become the operator brain.

The product is explicitly designed to stay advisory-only, preserve data honesty states, and avoid pretending proxy or fallback data is the same thing as execution-grade truth.

What is already real

• A real compact repo snapshot with frontend, backend, scripts, tests, and docs
• A browser-first operator desk structure
• An advisory-only commodity workflow
• AI Desk as a real product surface
• OpenAI integration and auth plumbing already present in the project
• Established oil, gold, and silver product direction
• A USOUSD trader-facing oil workflow with WTI held as research context
• Real signals, risk, chart, and paper-trade scaffolding
• Multiple trust and usability passes already implemented
• Commodities-first comparator and watchlist logic plus structured AI-advisor work landed in implementation passes

How the operating model works

The desk is designed around advisory workflow truth rather than fake platform certainty.

1. Market context enters through commodity-specific chart, signal, and risk surfaces instead of one generic cross-asset shell.

2. Oil, gold, and silver workflows stay grounded in explicit research framing, including USOUSD trader-facing context and WTI underlying context where relevant.

3. AI Desk sits on top of signals, charts, risk, and research surfaces as a synthesis and next-step layer rather than a magical autonomous trader.

4. Paper-trade proposals and operator guidance stay clearly separate from any claim of real-money execution.

5. Data freshness, live-versus-proxy distinctions, and top-shell clarity are treated as product behavior, not just backend details.

Why it matters

Most market dashboards either stay shallow or fake confidence. The more interesting systems problem is building a desk that helps an operator think clearly without blurring the line between research-grade context and execution-grade truth.

That is the actual strength here: not a claim that trading is automated, but an attempt to make charts, signals, risk, research, and AI guidance work together in a more honest and useful operator surface.

Current state

This is a strong pilot-stage system. The product direction is real, and the desk has already gone through multiple implementation and QA passes, but it should not be framed as a finished terminal replacement, a real-money execution platform, or a fully proven live AI workflow.

The main remaining work is about trust under live hydration, making AI Desk feel like a true operator brain instead of a structured wrapper, and tightening incomplete research and shell states before claiming stronger maturity.

What I would improve next

• Finish shell hydration reliability across top-shell surfaces
• Deepen AI Desk into a more stateful operator brain for commodities workflows
• Tighten research completeness and freshness signaling so desk context stays coherent under live conditions