Skip to main content

Component Breakdown

CIG is a monorepo containing specialized packages that work together to provide a seamless infrastructure intelligence experience.

Application Surfaces​

apps/landing​

  • Role: Public facing landing page.
  • Tech: Next.js.
  • Purpose: Product showcase, authentication entry point, and canonical logout return surface.

apps/dashboard​

  • Role: Primary user interface.
  • Tech: Next.js (App Router), Tailwind CSS.
  • Purpose: Infrastructure visualization, live/demo graph switching, graph exploration, agent chat, and settings management.

apps/wizard-ui​

  • Role: Guided installation.
  • Purpose: Helps new users set up their self-hosted environment and initial discovery jobs.

Core Packages​

@cig/api​

  • Role: Central Domain Server.
  • Core Tech: Fastify, GraphQL, WebSocket, Prometheus, Zod.
  • Responsibility: Orchestrates all logic between the UI, graph, discovery, semantic retrieval, and chat layers.

@cig/graph​

  • Role: Graph Engine.
  • Core Tech: Neo4j (Cypher), Neo4j-Driver.
  • Responsibility: Manages the graph schema, handles complex traversals, applies scope filters, and ensures data integrity.

@cig/discovery​

  • Role: Discovery Orchestrator.
  • Responsibility: Schedules discovery cycles, triggers the cartography service, and updates the graph state.

@cig/auth​

  • Role: Authentication & Session Hub.
  • Core Tech: jose, jsonwebtoken, Authentik.
  • Responsibility: Ensures secure access to the API and manages cross-platform user sessions for Authentik and Supabase-backed flows.

@cig/sdk​

  • Role: Typed Client Foundation.
  • Responsibility: A shared library providing typed API clients for the Dashboard and CLI, ensuring type safety throughout the codebase.

Intelligence & Agents​

@cig/agents​

  • Role: AI Reasoning Engine.
  • Core Tech: LangChain, OpenAI/Anthropic.
  • Responsibility: Implements the tool-calling logic that allows LLMs to query and refine the infrastructure graph.

@cig/chatbot​

  • Role: RAG Pipeline.
  • Responsibility: Manages vector ingest and retrieval to provide context to the agentic layer, including managed semantic collection scopes.

Infrastructure & Operator Tools​

@cig/iac​

  • Role: Infrastructure as Code.
  • Tech: Terraform.
  • Responsibility: Defines the AWS base infrastructure required to run CIG.

@cig/infra​

  • Role: Deployment Wrapper.
  • Tech: SST (Serverless Stack).
  • Responsibility: Manages the runtime delivery of containers to ECS/Fargate.

@cig-technology/cli​

  • Role: Operator Command Line.
  • Responsibility: Deployment, bootstrapping, environment management, demo provisioning, and terminal-based querying.

Services​

services/cartography (Python)​

  • Role: Raw Discovery Data Engine.
  • Tech: FastAPI, Python Cartography.
  • Responsibility: The worker service that connects to cloud APIs (AWS/GCP/Azure) and pushes raw data into Neo4j.

Vector Store (Chroma)​

  • Role: Specialized Vector Semantic Engine.
  • Responsibility: Powers the RAG pipeline by storing infrastructure metadata as high-dimensional embeddings for semantic retrieval. Managed deployments scope collections by tenant/workspace; demo mode uses the shared seeded demo workspace scope.