Seven agents debate a position over Telegram until they agree. Every number comes from a deterministic script; the language model only writes the argument.
Guided views
Explore this system
Step through curated paths without changing the source diagram.
Beat
Next
ReadyChapter 01 / 01
Guided chapter
Diagram guideExplore this system
Inspecting compiled semantics
E ExportT ThemeS Style0 Reset+ Zoom in- Zoom outEsc Close
Find a node
⌕/
No matching nodes
Semantic passport
Verified source
Authored reach
Route probeChoose a start node
Pick two semantic nodes on the diagram
Choose the source, then the destination. Direction matters.
Semantic lensCompare system roles
Choose up to two semantic kinds. One reveals its real traffic; two compare only direct authored relationships.
Choose a kind to inspect its nodes and touching relationships.
Semantic radar
Building overview
Click nodeDrag to pan
Debate
• Timo, Warren and Ray argue in turn and rebut each other
• Charlie audits every round for bias
• Stops on consensus or after four rounds
Numbers, not prose
• Quinn's conviction score is a Python function
• One cached data snapshot per discussion, so every agent sees identical figures
• Missing fair value falls back to on-chain MVRV
Learning loop
• Every verdict is journalled with its dissent
• Hit-rate scored against real prices at 7, 30 and 90 days
• Weak segments become proposals — a human approves each change
Architecture diagram • Built with Archify • Create yours ↗ • Hover to trace • R route • Click to focus • +/− zoom • M radar • T theme • E export