Play 50
Financial Risk Intelligence
Real-time market analysis, credit risk, fraud detection with explainable AI decisions.
Financial services AI agent — real-time market analysis, credit risk assessment, regulatory document processing (SEC/Basel III), fraud detection, with explainable AI decisions, audit trails, and human-in-the-loop escalation. Uses AI Search for regulatory knowledge, Event Hubs for market feed ingestion, and Cosmos DB for risk state management. Every decision includes confidence scores and reasoning chains.
Architecture Pattern
RAG-powered financial agent: real-time feeds, explainable decisions, human escalation
Azure Services
DevKit (.github Agentic OS)
- agent.md — root orchestrator with builder→reviewer→tuner handoffs
- 3 agents — Financial Risk Builder (gpt-4o), Reviewer (gpt-4o-mini), Tuner (gpt-4o-mini)
- 3 skills — deploy (251 lines), evaluate (163 lines), tune (213 lines)
- 4 prompts — /deploy, /test, /review, /evaluate with agent routing
- .vscode/mcp.json — FrootAI MCP with OpenAI + Cosmos DB inputs + envFile
TuneKit (AI Config)
- config/openai.json — gpt-4o, temp=0.1, deterministic
- config/risk.json — risk models, confidence thresholds, escalation rules
- config/guardrails.json — PII protection, regulatory compliance
- evaluation/eval.py — Risk accuracy >90%, Fraud detection >95%
Tuning Parameters
Machine evidence
FrootAI evidence lifecycle
This is an internal evidence maturity label, not third-party certification, accreditation, legal compliance, or a production guarantee. Missing or expired evidence demotes automatically; catalog claims cannot promote a play.
This play currently has design evidence only. A runnable scenario, endpoint evaluation, and build receipts are the next contiguous gates.
Repo Intelligence
v1A no-clone, revision-pinned map for agents and humans. Observed evidence is separated from inferred flow so the output stays useful without pretending to be a full call graph.