How iiooiioo888/linkin works
Linkin is an AI runtime and monitoring console. A user enters a query in the chat UI, can request a routing preview, then sends it through the chat API and shared execution-path decision: simple requests use the reflective generation loop, while company requests are analyzed into milestones and delegated to seats. The repository also documents integrations and memory recall, OPC sensing with guarded writes, AI Hub routing, and virtual-credit billing. The workflow edges follow the README and shown excerpts; unsampled subsystem internals are represented at their existing directory paths.
Main entry points: graph.py Execution graph · routes.py Chat endpoints · tools.py Agent tools · ChatView.tsx Chat UI
Components 21
Chat interaction
- Chat UI
frontend/src/components/ChatView.tsx - Route preview
frontend/src/hooks/useRoutingPreview.ts - API client
frontend/src/api/client/index.ts - FastAPI app
backend/main.py - Chat endpoints
backend/api/routers/chat/routes.py
Unified runtime
- Route decision
backend/core/execution_path.py - Execution graph
backend/core/graph.py - Reflective loop
backend/core/nodes/reflection.py - Memory retrieval
backend/memory/vector_store.py - Integration recall
backend/integrations/recall_bridge.py
Company execution
- Task manager
backend/services/task_manager/runtime.py - Seat orchestration
backend/company/orchestrator/execution/run.py - Agent tools
backend/company/tools.py
Platform services
- OPC sensing
backend/core/opc_degrade.py - OPC UA service
opc_service/app.py - Write guardrail
services/guardrail-service/app.py - AI Hub
backend/hub/router.py - Credits and billing
backend/billing/pricing_engine.py
Monitoring console
- Monitoring console
frontend/src/components/AppShell.tsx - Monitoring services
backend/services/agent_monitor/aggregate.py
Other components
- User
Connections 21
- User → Chat UI: submits query
README.md - Chat UI → Route preview: requests preview
README.md - Route preview → API client: calls preview API
frontend/src/hooks/useRoutingPreview.ts - API client → Chat endpoints: sends chat request
README.md - FastAPI app → Chat endpoints: mounts routes
backend/main.py - Chat endpoints → Route decision: builds preview
backend/api/routers/chat/routes.py - Chat endpoints → Execution graph: invokes graph
backend/api/routers/chat/routes.py - Route decision → Execution graph: selects execution path
README.md - Execution graph → Reflective loop: runs simple path
README.md - Execution graph → Task manager: runs company path
README.md - Execution graph → Memory retrieval: retrieves and saves
README.md - Execution graph → OPC sensing: adds sensor context
README.md - Execution graph → Integration recall: adds recall context
README.md - Task manager → Seat orchestration: delegates seat work
README.md - Seat orchestration → Agent tools: uses agent tools
- OPC sensing → OPC UA service: reads sensor data
- Agent tools → Write guardrail: guards writes
README.md - Chat endpoints → AI Hub: routes model requests
- Agent tools → Credits and billing: meters tool use
backend/company/tools.py - User → Monitoring console: views status
README.md - Monitoring console → Monitoring services: requests monitoring