Operator Agent System

Private-access Operator workspace where five specialist AI perspectives organize market-structure labels, hypothetical risk arithmetic, reflective prompts, model-factor summaries, and technical-model education. The tier combines these perspectives with a seven-engine synthesis for independent research review. It is educational financial-data software, not a professional trading desk, adviser, signal provider, or execution service.

What this page covers

• Who the 5 specialist agents are and what each one does
• Which agent perspective to use for each part of a research review
• The real methodologies driving each agent (ICT, Kelly, CBT, confluence scoring, OTE)
• How to navigate to each agent and what questions to ask

The Five Specialist Agents

The Operator Agent system is not a single AI — it is five distinct specialist AIs, each trained on a different methodology and given a different personality, a different scope of responsibility, and a different set of rules about what they will and won't do.

The perspectives are separated so structure labels, arithmetic examples, reflection prompts, factor summaries, and technical terminology remain distinguishable. Users must independently verify every output; no perspective authorizes a transaction.

Which Agent for Which Stage

How to Access Each Agent

Each agent has their own dedicated page in the app, reachable from the sidebar or directly by URL. The agents are fully conversational — ask them questions in natural language exactly as you would a specialist mentor.

1. Navigate to an agent page

   Each agent lives at /agents/[name] — e.g. /agents/marco, /agents/rex, /agents/jade, /agents/nova, /agents/kai. The sidebar under 'Operator Agents' links all five.

2. Ask a focused educational question

   Ask about a concept, research method, or clearly fictional example. For example: 'How can a student compare two possible D1 structure labels and annotate an FVG on a historical chart?'

3. Use the sample questions

   Each agent page shows starter questions that demonstrate their specialty. These are calibrated to produce the most informative responses — use them as templates for your own questions.

4. Build a sequential study workflow

   For a structured paper exercise, compare Marco's structure labels, Rex's arithmetic example, Nova's bull-and-bear factor comparison, Kai's execution-model explanation, and Jade's reflection prompts. The sequence does not constitute a professional service or trade workflow.

Agent Scope Reference

The Agent Rule System

Each agent has hard-coded rules that define what they will and will not engage with. These aren't suggestions — they're enforced at the system level to maintain specialist integrity.

• Marco will never predict specific price targets — only map structure and alternative readings
• Rex explains risk arithmetic but does not determine personal exposure
• Jade will never dismiss an emotional response — she treats all states as data
• Nova compares bull and bear model factors without selecting a preferred outcome
• Kai explains execution-model concepts but does not identify a valid live entry

How the 7-Engine Synthesis Works

When you activate an Operator synthesis on a single pair, seven analytical engines produce separate model readings: Wyckoff phase, volume anomaly, session timing, pattern classification, liquidity-sweep classification, divergence, and momentum. The coordination layer records agreement and conflict without treating either as probability or approval.

The output is a structured educational briefing comparing bull and bear scenario labels, pattern and sweep observations, volume context, session context, and hypothetical risk arithmetic. Every section requires independent review and may be inaccurate; it does not select a final outcome.

Fictional Multi-Agent Study: XAUUSD London Example

This fictional paper exercise shows how five agent perspectives can be compared. It is not a live trade, individualized analysis, or execution workflow.

1. Step 1 — Marco: Compare HTF Labels

   Provide a fictional chart annotation and ask Marco to compare possible W1, D1, and H4 structure labels. Record disagreements and verify them against the source chart.

2. Step 2 — Rex: Review Hypothetical Arithmetic

   Provide fictional account, distance, and correlation inputs and ask Rex to show the arithmetic. Verify the calculation independently; the result is not a personal position-size recommendation.

3. Step 3 — Nova: Compare Model Factors

   Provide fictional structure, volume, and session labels. Nova returns a model-factor summary without approving or rejecting a transaction.

4. Step 4 — Kai: Explain an OTE Example

   Provide fictional FVG and displacement levels and ask Kai to explain how an OTE reference would be calculated. Record the example for retrospective study; no live order, stop, partial, or runner instruction is implied.

5. Step 5 — Jade: Reflect on a Fictional Decision Log

   Provide a fictional journal note and ask Jade for non-clinical reflection questions. The response does not evaluate a live setup, provide a protocol, or replace a qualified professional.

OPERATOR SYNTHESIS | PRIVATE STUDY SAMPLE
═══════════════════════════════════════════════════════

ACCESS: PRIVATE OPERATOR WORKSPACE
SCENARIO COMPARISON: BULL CASE ↔ BEAR CASE
No preferred outcome is selected.

── WYCKOFF MODEL ────────────────────────────────────
Bull case: accumulation-style phase label is possible
Bear case: distribution-style phase label is also possible
Review note: compare both labels with the source chart

── VOLUME MODEL ─────────────────────────────────────
Comparison: activity label is elevated in the sample
Review note: volume does not establish participant identity or direction

── SESSION CONTEXT ──────────────────────────────────
Configured study window: a named session
Review note: timing does not establish probability or suitability

── MODEL COMPARISON ─────────────────────────────────
Bull evidence: structure and context labels recorded
Bear evidence: rejection and distribution labels recorded
Conflict: assumptions and alternative interpretation remain documented

STATUS: STUDY SAMPLE — NO FINAL VERDICT
NO ENTRY, TARGET, STOP, PRICE, LOT SIZE, OR EXECUTION INSTRUCTION

Frequently Asked Questions