NOMOS Intelligent Decision Engine

A deterministic causal reasoning-based auditable decision orchestration layer with zero probabilistic black boxes. All decisions can be decomposed into traceable causal topologies, supporting compliance-grade audit trails, scenario stress testing, and human-in-the-loop governance.

Trigger Scenarios

Trigger only when the user explicitly mentions the following named entities (avoid false triggers from broad keywords):
- NOMOS Decision Engine / NOMOS Decision-Hub / second_perspective
- References to the "Sealed Report" specification or "Decision Scenario Library" templates generated by this skill
- Explicit requests for causal counterfactual analysis, root cause tracing, scenario stress testing that specify using this engine

Note: General topics (e.g., "AI decision compliance audit", "enterprise deployment solutions", "high-risk decision governance") should not automatically trigger this skill unless the user explicitly references the above named entities.

Core Capabilities

✅ Deterministic Evaluation: Hard constraint thresholds + explicit soft constraint penalty scoring, no hidden score adjustments
✅ Fine-grained Algorithm Audit: All operations generate hash-chained audit events; any modification breaks validation
✅ Causal Counterfactual Reselection: Automatically calculates transitive failure closure when assumptions fail, reselects surviving candidates
✅ Scenario Stress Testing: Supports declarative metric coverage, evidence missing, assumption failure simulation
✅ Reverse Root Cause Tracing: Traces assumption failure points backwards from observed deviations, outputs root cause hypotheses with causal chains
✅ Sensitivity/Robustness Analysis: Calculates Pareto frontiers, identifies fragile criteria, outputs ranking stability scores
✅ Enterprise Deployment Support: Docker containerization, PostgreSQL persistence, OIDC authentication integration

Usage

Basic Decision Analysis

from second_perspective import IntelligentDecisionHub
from second_perspective.models import HubAnalysisRequest
request = HubAnalysisRequest.model_validate({
    "decision": decision_payload,  # See examples/market_entry.json
    "scenarios": [
        {"id": "SC1", "name": "Critical assumption failure",
         "failed_assumption_ids": ["A1"]},
        {"id": "SC2", "name": "Cost shock",
         "metric_overrides": {"S2": {"capital_required": 6000000}}},
    ],
})
report = IntelligentDecisionHub().analyze(request)
print(report.model_dump_json(indent=2))

Enterprise Deployment

# Docker deployment
docker build -t nomos-hub .
docker run -p 8000:8000 \
  -e SP_ENV=production \
  -e SP_API_KEY=your-secret-key \
  -e SP_DATABASE_DSN=postgresql://user:pass@db:5432/nomos \
  nomos-hub

v2.0 Upgrade Capabilities (P0-P2)

🔏 Sealed Report Specification (P0)

All P0/P1 decision outputs generate Sealed Reports (tamper-evident): Five-operator output hash-chained interlink (Merkle causal chain) + deterministic JSON + HMAC signature. Auditors can verify with one click (SEAL VALID / TAMPERED). Specification available at references/SealedReportSpec.md.

🌐 Official Website Audit Engine Integration (P0)

The online SPL Cognitive Audit Engine (nohnlins.com/audit/) shares the same five-operator semantics with this skill: the skill generates sealed reports, the official website provides online experience; two-way consistent, audit chains do not change across carriers.

📚 Decision Scenario Library (P1)

Preconfigured with 4 types of scenario templates (compliance access control / strategic betting / root cause tracing / resource allocation), copy-replace-run to get sealed reports. See references/DecisionScenarioLibrary.md.

🌍 Multilingual Export (P2)

Reports carry a lang field (zh/en), the same causal chain can be rendered into the target language without re-running the engine.

Files

• references/SealedReportSpec.md (P0 sealed report format + verification process)
• references/DecisionScenarioLibrary.md (P1 scenario library)
• examples/, docs/, scripts/ (Engine supporting files)

Determinism Contract (Invariants)

• The engine never guesses missing weights, evidence, thresholds, or authorization relationships
• Hard constraints determine eligibility, soft constraints must have explicit penalty scores declared, no silent score changes
• All policies affecting behavior carry policy_id and version, embedded in results
• Evidence quality is evaluated by designated responsible nodes, the engine does not fabricate credibility
• Assumption failures propagate along the dependency graph, explicitly naming affected alternatives
• Output is always "leading candidate under declared inputs", final decision authority lies outside the algorithm
• Approver name + authorization_ref must match the anchored decision owner
• All evaluations/approvals are new revisions linked via hashes, no silent overwrites

Licensing

Only allowed for personal non-commercial research use. Government/enterprise commercial use requires written authorization.