Layer 02 · CONNECT · four-layer stack

Connect

A swarm, not a lone model — 54 agents, one substrate.

Correlation is where raw signal becomes intelligence. Events gain meaning when linked to entities, campaigns and prior knowledge. That work is done by a swarm of specialised agents — each with a narrow competence, together covering a wide surface. The registry is real: 54 agents, machine-readable.

See the live graph/api/swarm →
L2 · PROOFRegistry composition
Agents in registry
real topology
Registry edges
co-membership + hubs
Domains
strategy → operations
Honeypot events
what they correlate
L2 · HOWHow the swarm works
A

Specialisation is the point

Agents are categorised by domain (strategy, intelligence, security, engineering, operations) and tier (fast, smart, deep). A fast-tier agent triages; a deep-tier agent adjudicates. Routing by competence beats one model doing everything mediocrely.

B

Topology you can check

The graph on /swarm.html is not decoration — nodes and edges come from the real agent registry (domain co-membership plus cross-domain hub links), served machine-readable from /api/swarm. Click a node to read what the agent actually does.

C

Threads over chat

Agents coordinate through threads bound to root events, with summaries and status — shared memory rather than ephemeral chat. Outputs are structured so downstream agents (and auditors) consume them without re-reading prose.

D

Agent-native by design

The whole platform speaks to machines: a well-known manifest at /.well-known/agent-manifest.json, llms.txt for LLM readers, and JSON endpoints for every published figure. Agents are first-class readers of this site.

L2 · NEXTWhere correlation goes
→ L3

Reasoning layer

Correlated signal becomes judgement in OpenMythos.

→ layer 03 · reason
→ visual

Live swarm graph

Interactive force-directed real topology with D1 liveness.

→ /swarm
→ register

Agent registry

Full filterable registry with manifests and run history.

→ /agents