Defensive engagement & threat intelligence research laboratory. Converts inbound scam emails into actionable IOCs through controlled, policy-driven AI engagement. Multi-agent LLM architecture with adaptive strategy selection. Docs-only preview.

Structured refinement of any document between AI agents. Two agent roles bounce a file back and forth using disagreement markers until it converges. Open source, agent-agnostic, runs from any terminal.

RECODE: REquirements recovery from CODE for adoption review of autonomous driving software

Hoshimiya Script / StarPolaris OS — internal multi-layer AI architecture for LLMs. Self-contained behavioral OS (Type-G Trinity).

The Verifier Is All You Need: Six Architectural Interventions That Don't Matter and the Verification Boundary That Does

Reinforcement-learning aggregator on top of the TradingAgents multi-agent LLM trading framework. PPO policy that beats buy-and-hold on the 2026 YTD test across two LLM backbones (Anthropic + OpenAI); cross-LLM transfer holds. Course project, Columbia IEORE 4733.

🛡️ Transform scam emails into actionable threat intelligence with ScamBuster, enabling defensive security and fraud prevention through controlled engagement.