How yadavallicharishma3-art/vidhura works
VIDHURA is an incident-response cockpit: an SRE submits a simulated incident, the backend recalls prior experience, generates evidence-first investigation guidance, and lets the engineer save a confirmed outcome for future recall. The graph separates the UI/API, orchestration, memory, and reasoning responsibilities. The README documents major capabilities such as memory browsing, dashboard/status, and seeding; some UI and route details are represented at subsystem level because their implementation was not fully sampled.
Main entry points: main.py FastAPI routes · hindsight_service.py Memory adapter · agent.py Incident agent · seed_memories.py Memory seeding
Components 15
Incident experience
- Incident cockpit
frontend/src/App.tsx - React entry
frontend/src/main.tsx - Curated scenarios
data/incidents.json
Backend workflow
- FastAPI routes
backend/main.py - Request models
backend/models.py - Incident agent
backend/agent.py - Incident records
backend/incident_service.py - Provider configuration
backend/config.py
Memory services
- Memory seeding
scripts/seed_memories.py - Memory adapter
backend/hindsight_service.py - Hindsight service
- Demo memory store
backend/hindsight_service.py
Reasoning providers
- Investigation reasoning
backend/llm_service.py - Groq GPT-OSS
Other components
- SRE engineer
Connections 21
- SRE engineer → Incident cockpit: submits incident
README.md - React entry → Incident cockpit: renders
frontend/src/main.tsx - Incident cockpit → FastAPI routes: calls API
frontend/src/App.tsx - FastAPI routes → Request models: validates input
backend/main.py - FastAPI routes → Incident agent: runs investigation
backend/main.py - Incident agent → Incident records: creates and updates
backend/agent.py - Incident agent → Memory adapter: recalls experience
backend/agent.py - Incident agent → Investigation reasoning: requests guidance
backend/agent.py - Investigation reasoning → Groq GPT-OSS: requests reasoning
backend/llm_service.py - FastAPI routes → Memory adapter: stores feedback
backend/main.py - SRE engineer → FastAPI routes: confirms outcome
README.md - Memory adapter → Hindsight service: recalls and retains
backend/hindsight_service.py - Memory adapter → Demo memory store: reads and writes
backend/hindsight_service.py - Memory adapter → Provider configuration: reads settings
backend/hindsight_service.py - Investigation reasoning → Provider configuration: reads settings
backend/llm_service.py - Incident records → Provider configuration: uses data paths
backend/incident_service.py - FastAPI routes → Memory adapter: lists memory and status
backend/main.py - FastAPI routes → Memory seeding: starts seeding
backend/main.py - Memory seeding → Curated scenarios: reads scenarios
scripts/seed_memories.py - Memory seeding → Memory adapter: retains scenarios
scripts/seed_memories.py - FastAPI routes → Incident records: lists and fetches incidents
backend/main.py