Decision Augmentation Engine

⌨️ Stage 0 — User Input

Raw Problem Statement

User enters what they believe the problem is. Unstructured, unvalidated, potentially mis-specified.

Context

Actor, geography, objective, constraints, timeline

Stakes

Complexity flag → routes to advisor or guru layer

Budget

Daily limit, per-query cap, advisor tier preference

↓

raw query

🎯 Stage 1 — Problem Specification Engine

Corpus Check

Pattern Matching

Does this problem match known problem types in the corpus? What domain? What strategic school applies?

Live Context

Fact Grounding

Current polling, news, sentiment data. Is the problem statement consistent with current facts?

Formalisation

Structured Spec

Actor · Audience · Objective · Constraints · Success metric · Problem type classification

↕

iterate until approved

✅ User Approves Refined Problem Specification

The most important gate in the system. No analysis proceeds until the problem is precisely specified.

↓

approved spec

⚔️ Stage 2 — Domain Debate Engine

🏛️

Power & Institutions

Machiavelli, Kissinger, Clausewitz

📣

Movement Building

Alinsky, Tarrow, Ostrom

🖼️

Media & Framing

Lakoff, Cialdini, Bernays

🗳️

Electoral Strategy

Targeting, mobilisation, persuasion

🔬

Behavioural Economics

Kahneman, Thaler, nudge theory

🌐

Geopolitical Realism

Mearsheimer, Waltz, structural forces

Each school retrieves its most relevant corpus chunks → generates competing strategy + tactics → outputs structured position with explicit tradeoffs

↓

N competing positions

🎓 Stage 3 — Advisory Layer · Sonnet-Powered

🧠

Political Strategist

Trained Sonnet persona

~$0.50–2.00 / query

📊

Behavioural Economist

Trained Sonnet persona

~$0.50–2.00 / query

📡

Communications Expert

Trained Sonnet persona

~$0.50–2.00 / query

⚖️

Policy Analyst

Trained Sonnet persona

~$0.50–2.00 / query

🔮

+ Add Advisor

User selects which advisors to consult

expandable

Each advisor adjudicates the debate from their trained perspective → returns recommendation + explanation + confidence score

↕

escalate if complex

🔭 Stage 4 — Guru Layer · Opus-Powered · On Demand

🏆

Domain Expert Alpha

e.g. Electoral systems, comparative politics

~$5–20 / query + comms attribution share

🏆

Domain Expert Beta

e.g. Behavioural finance, risk psychology

~$5–20 / query + comms attribution share

🏆

Domain Expert Gamma

e.g. Media strategy, narrative architecture

~$5–20 / query + comms attribution share

Heavily trained Opus personas · deep reasoning · long context · explicit escalation by user · highest cost, highest depth

↓

adopted strategy

📤 Stage 5 — Integrated Communications

Campaign Emails

Strategy informs tone, framing, sequence. Case ID embedded for attribution tracking.

Social Media

Paid + Organic

Ad creative, messaging hierarchy, platform targeting — all derived from adopted strategy.

Website / Portals

Landing Pages

Submission portals, campaign pages — built and deployed from strategy output.

Attribution

Case ID Tracking

Every comms output carries a case ID → tracks downstream activity → flows back to compensation engine.

↓

outcomes observed

↩ Stage 6 — Feedback Loop & Probabilistic Engine

📈

Polling Data

Live tracking fed back into corpus. Did sentiment move as predicted?

🌐

Social Monitoring

Engagement, reach, narrative adoption — real-world signal vs model prediction.

✅

Outcome Validation

Users and contributors report actual outcomes. Closes the loop. Most valuable data layer.

🎲

Probabilistic Weights

Bayesian updating of school and advisor weights based on observed accuracy over time.

Feedback flows back to Stage 1 (live context), Stage 2 (school weights), Stage 3 (advisor accuracy scores), and Stage 4 (guru performance)

↓

predictive layer

🔮 Stage 7 — Sentiment Trajectory & Prediction Engine

Time-Series Models

Sentiment Trajectory

Given current data, where does this issue go? Rate of formation, decay, peak timing.

Outcome Probability

Decision Outcomes

P(success) for each strategy option given current context. Confidence intervals widen or narrow with more case data.

Strategy Adjustment

Real-Time Guidance

"Based on current trajectory, your strategy should shift emphasis here." Active rather than static advice.

↓

$0.12

This Query

$4.80

Today

$25.00

Daily Limit

12,400

Tokens Used

Advisor

Current Tier

$1.40

Advisor Earnings