Execution mode
Global switch — hard-coded off in this build
Paper / shadow mode only. Signals, sizing and journaling run end to end; the order-placement methods on both adapters reject by design. Enabling live routing is a deliberate, separate step that requires server-side credentials and an approval gate.
Service boundaries
Prepared interfaces — nothing is connected
Kraken Derivatives
REST + WebSocketnot configuredFutures engine · MNQU6 · MESU6 · MBTQ6
Boundary is defined and mocked. No credentials stored, no socket opened, no orders routed.
Robinhood Agentic / MCP
Agentic / MCPnot configuredOptions engine · long CALL / long PUT only
Contract-only adapter. Multi-leg, spreads, and short premium are out of scope by design.
Credential policy
Secrets never reach the browser
- · No API key, secret or token exists in frontend code or bundled assets.
- · When brokers are connected, keys live in project secrets and are read only inside server functions.
- · The browser calls typed server functions; it never talks to a broker directly.
- · Request signing, nonces and WebSocket auth stay server-side behind the adapter interfaces.
- · This build persists nothing: everything on screen is local mock data.
Risk defaults
Edited here, enforced by the risk engine for both engines
Changes are local to this session in mock mode
Methodology and attribution
The options quality layer is an independent, original checklist inspired by widely discussed company-quality-and-thesis-first investing principles. It does not reproduce, replicate, or represent any third party's proprietary system, course material, or signals, and is not affiliated with or endorsed by any such party. ICT concepts are implemented as this project's own interpretation of publicly described price-action ideas.
Risk disclaimer. This is a personal research and journaling tool, not investment advice, and not a broker. Futures and options trading involves substantial risk of loss and is not suitable for every investor; you can lose more than your initial outlay. All data shown here is simulated mock data for interface development. Scores, theses, and AI commentary are informational only, never recommendations, and never override your own rules or the risk engine. Past performance does not indicate future results.