Logos - alexar76/aicom GitHub Wiki
LOGOS โ federation analytics
LOGOS is the ecosystem's read-only federation analytics engine. It takes real Hub snapshots and measured settlement volume, runs rolling z-score anomaly detection, correlates across sources, and answers natural-language questions through Metis โ without inventing zeros for missing data.
Live dashboard: logos.modelmarket.dev ยท Landing: alexar76.github.io/logos ยท Source: github.com/alexar76/logos
Why LOGOS matters in AICOM
| Layer | Role |
|---|---|
| Hub / peers | Live catalog, invoke routes, settlement telemetry |
| MOMUS / SKOPOS / Treasury | Security findings, fleet posture, bounty pay โ polled as sources |
| LOGOS | Observatory โ snapshots, anomalies, cross-source insights, guarded assistant |
| Metis | Optional NL reasoning behind insights (fail-closed when unreachable) |
Hard invariant: LOGOS is read-only by construction. It does not scan, remediate, pay, or deploy. An unreachable source stays unavailable / โ, never a fake 0.
What operators get
- Real Hub snapshots + measured settlement volume (not projected)
- Rolling z-score anomalies and cross-source correlations
- Natural-language insights (Metis) + a guarded chat surface
- 5-language UI; SQLite or Postgres store
Quick start
git clone https://github.com/alexar76/logos.git && cd logos
pip install -e ".[dev]"
LOGOS_HUB_URL=https://modelmarket.dev python -m logos.main
# Frontend: cd frontend && npm install && npm run dev
Monorepo path: logos/ in alexar76/aicom (factory mirror excludes satellites).
Related
- Ecosystem โ monorepo map
- Metis โ cognition tier used for NL insights
- Oracles โ math providers LOGOS may observe via Hub
- Knowledge base:
docs/ecosystem/knowledge-base.md