Info only · Not financial advice · Conservative bias

AI market-intelligence site for Sugar SB N26.

A compact command center that checks credible sugar-market drivers, separates facts from assumptions, scores directional pressure, and produces 12h, 24h, and 48h advisory views with confidence levels.

Run Intelligence Brief System Status
Control Center

Run the SB N26 intelligence brief

Click "Run Live AI Research" to let OpenAI GPT research fresh market data and generate the brief automatically, or enter values manually.

Market Inputs

Use verified data only. Unknown fields are allowed but reduce confidence.

Manual / API-ready

Live AI mode: GPT researches fresh SB N26 prices, news, supply-demand, weather, and macro signals via web search, then fills this dashboard automatically. A full run can take 1–5 minutes.

Advisory Output

Waiting for fresh inputs.

Info only
Directional Score —
Advisory View No brief yet
Data Quality —
12h Pending

Run the brief to generate near-term direction.

Confidence: —
24h Pending

Run the brief to generate tactical direction.

Confidence: —
48h Pending

Run the brief to generate short-cycle direction.

Confidence: —

Compact Report

Generate a brief to see the structured market report here.
Credible Source Room

Sources the agent should check every run

Use these as the primary source universe. Live mode should fetch, timestamp, and score recency before producing output.

Decision Engine

Forecast logic and confidence method

1. Facts First

Collect latest price, daily change, support/resistance, volume/open interest, and verified news. Anything not timestamped is treated as lower quality.

2. Driver Scores

Convert technicals, news, Brazil supply, India policy, Thailand crop, weather, crude, USD, and risk sentiment into bearish, neutral, or bullish pressure.

3. Horizon Weighting

12h favors technicals and market sentiment. 24h blends macro and supply. 48h gives more weight to policy, weather, and supply-demand headlines.

4. Confidence Control

Confidence rises when data is fresh and drivers agree. It drops when sources conflict, fields are missing, or the directional score is too close to neutral.

Scoring Weights

Factor12h24h48h
Technicals30%20%12%
News sentiment22%20%16%
Supply / policy18%28%34%
Weather risk10%14%20%
Crude / ethanol parity10%10%8%
USD / broad risk10%8%10%
Report Template

Agent output structure

Required Report Sections

  1. Agent Name
  2. Purpose
  3. Data Sources
  4. Market Factors Checked
  5. Forecast Logic
  6. Confidence Scoring Model
  7. Output Report Template
  8. Example Sugar SB N26 Report
  9. Implementation Notes
  10. Risk Disclaimer

High-Risk Self-Check

  • Is the SB N26 price fresh?
  • Are sources credible and timestamped?
  • Are headlines recent enough to matter?
  • Is confidence justified by data quality?
  • Is advice clearly labeled info-only?
  • Are 12h, 24h, and 48h calls different where needed?
Live System Status

This system is live

SugarPulse is deployed in production on MJ's cloud server with a secure backend — the OpenAI key never touches the browser. Every “Run Live AI Research” click triggers the full 3-call GPT-5.5 pipeline below.

✓

Hosting — LIVE

Served from MJ's cloud computer on port 8090 under pm2 process management, with auto-restart on reboot and firewall rules in place.

1

Call 1 · Technical research — LIVE

GPT-5.5 with live web search collects verified price, change, volume, open interest, and published support/resistance. Unverified values stay blank — never guessed.

2

Call 2 · News research — LIVE

GPT-5.5 with live web search gathers credible, dated headlines plus Brazil, India, Thailand, weather, and macro facts. Runs in parallel with Call 1.

3

Call 3 · Analysis — LIVE

GPT-5.5 combines only the verified facts from Calls 1 and 2 into factors, 12h/24h/48h forecasts, and the full brief — no assumptions, facts only.