How aastik0303/orchestrator_system works
This is a runnable multi-agent chat starter: a user submits a message, optional files, and an agent choice; the chat API persists the conversation, invokes the LangGraph workflow, and returns the selected worker’s response. The workflow routes, validates, retries, and falls back. The repository also includes a React dashboard with workspace and operational pages. Several worker capabilities are scaffolded; frontend-to-backend wiring and some worker internals were not shown, so uncertain connections are omitted.
Main entry points: langgraph_flow.py LangGraph workflow · supervisor.py Agent registry · chat.py Chat and uploads · App.tsx App shell
Components 24
Web client
- App shell
frontend/src/App.tsx - Chat workspace
frontend/src/components/workspace/WorkspacePage.tsx - Dashboard
frontend/src/components/dashboard/DashboardPage.tsx - Operations pages
frontend/src/pages - API client
frontend/src/services/apiClient.ts
Chat API and state
- FastAPI app
backend/app/main.py - Chat and uploads
backend/app/routes/chat.py - Chat session store
backend/app/services/chat_store.py - Runtime snapshots
backend/app/routes/chat.py
Orchestration
- Request and response models
backend/app/models.py - LangGraph workflow
backend/app/orchestrator/langgraph_flow.py - Agent selection
backend/app/orchestrator/router.py - Agent registry
backend/app/orchestrator/supervisor.py - Response validation
backend/app/orchestrator/langgraph_flow.py - Retry and fallback
backend/app/orchestrator/langgraph_flow.py - Workflow checkpointer
backend/app/orchestrator/langgraph_flow.py
Worker agents
- Chat and research
backend/app/agents/general_chat.py - Document RAG
backend/app/agents/document_rag.py - YouTube RAG
backend/app/agents/youtube_rag.py - Code development
backend/app/agents/code_dev.py - Data analysis
backend/app/agents/data_analyst.py - Groq client
backend/app/services/groq_client.py
Other components
- Chat user
- Groq API
Connections 28
- Chat user → App shell: uses
README.md - App shell → Chat workspace: routes to
frontend/src/App.tsx - App shell → Dashboard: routes to
frontend/src/App.tsx - App shell → Operations pages: routes to
frontend/src/App.tsx - FastAPI app → Chat and uploads: registers
backend/app/main.py - FastAPI app → LangGraph workflow: initializes
backend/app/main.py - FastAPI app → Chat session store: initializes
backend/app/main.py - Chat and uploads → Chat session store: reads and writes
backend/app/routes/chat.py - Chat and uploads → LangGraph workflow: invokes
backend/app/routes/chat.py - Chat and uploads → Request and response models: constructs
backend/app/routes/chat.py - Chat and uploads → Runtime snapshots: records runtime data
backend/app/routes/chat.py - LangGraph workflow → Request and response models: processes
backend/app/orchestrator/langgraph_flow.py - LangGraph workflow → Agent selection: selects agent
backend/app/orchestrator/langgraph_flow.py - LangGraph workflow → Agent registry: dispatches through
backend/app/orchestrator/langgraph_flow.py - LangGraph workflow → Response validation: validates response
backend/app/orchestrator/langgraph_flow.py - Response validation → Retry and fallback: routes invalid response
backend/app/orchestrator/langgraph_flow.py - LangGraph workflow → Workflow checkpointer: persists workflow state
backend/app/orchestrator/langgraph_flow.py - Agent registry → Chat and research: registers worker
backend/app/orchestrator/supervisor.py - Agent registry → Document RAG: registers worker
backend/app/orchestrator/supervisor.py - Agent registry → YouTube RAG: registers worker
backend/app/orchestrator/supervisor.py - Agent registry → Code development: registers worker
backend/app/orchestrator/supervisor.py - Agent registry → Data analysis: registers worker
backend/app/orchestrator/supervisor.py - Document RAG → Groq client: requests completion
backend/app/agents/document_rag.py - Chat and research → Groq client: uses completion
backend/app/agents/general_chat.py - Code development → Groq client: uses completion
backend/app/agents/code_dev.py - Groq client → Groq API: calls
backend/app/services/groq_client.py - Chat workspace → API client: uses
- API client → Chat and uploads: sends chat requests