How supremeai-bd/supremeai works
SupremeAI is a single-door monorepo for Core, Gateway, Browser relay, and the MCP Control Plane. A documented consumer workflow reaches the OpenAI-compatible Gateway surface; the README identifies the active service areas but does not establish their internal runtime wiring. The graph therefore shows that external entry point and the documented service topology, while omitting speculative calls. The source index identifies several Core modules, but some listed paths are absent from the supplied tree, so they are not represented.
Main entry points: dynamic_llm_router.py Gateway surface
Components 21
API Gateway
- Gateway surface
gateway/dynamic_llm_router.py - Gateway persistence
gateway/service/storage.py
Core Platform
- Core API
core/api/server.py - Core application
core/app.py - Model routing
core/brain/model_router.py - Task execution
core/brain/task_execution_engine.py - Adaptive tasks
core/adaptive_engine/task_engine.py - Self evolution
core/self_evolution/evolution_engine.py - Agent evolution
core/tier8/agent_evolution_engine.py - Governance policy
core/governance/policy_engine.py - Error reporting
core/errors/error_bus.py - Redis cache
core/cache/redis_manager.py - Supabase storage
core/database/supabase_client.py
Browser Relay
- Browser relay
browser/server.py - Browser sessions
browser/session_manager.py - Browser routes
browser/routes/_scraping.py
MCP Control Plane
- MCP task engine
infrastructure/mcp-control-plane/src/tasks/engine.ts - MCP guardian
infrastructure/mcp-control-plane/src/guardian/engine.ts - MCP risk policy
infrastructure/mcp-control-plane/src/policy/risk.engine.ts - MCP remediation
infrastructure/mcp-control-plane/src/remediation/engine.ts
Other components
- API consumer
Connections 1
- API consumer → Gateway surface: calls API
README.md