Skip to content

Extensions

Extensions are optional add-ons for organisations with specific needs. They are not required for core Fawkes functionality.

What Is Core vs Extension?

Core Fawkes provides everything needed to run a production-grade Internal Developer Platform out of the box:

  • GitOps delivery with ArgoCD
  • Developer portal with Backstage
  • CI/CD pipelines with Jenkins
  • Observability (Prometheus, Grafana, OpenSearch, Tempo)
  • Security scanning (SonarQube, Trivy, Vault)
  • Collaboration (Mattermost, Focalboard)
  • DORA metrics collection and dashboards
  • Dojo learning environment

Extensions add advanced capabilities that come with meaningful operational overhead. They are designed for organisations that have already stabilised the core platform and have a specific need.

Available Extensions

Data Platform

Directory: extensions/data-platform/

Adds data cataloging, lineage tracking, and data quality validation.

Component Purpose
DataHub Enterprise data catalog — discover, document, and track lineage across datasets
Great Expectations Data quality validation with Prometheus metrics export

When to add: Your organisation manages multiple data sources and needs a catalog, or you have data quality requirements that need automated validation.

Resource cost: 4–8 GB RAM, 4 vCPU additional. Requires OpenSearch (Tier 2).

the Data Platform Extension (available in the repository under extensions/data-platform/)

AI

Directory: extensions/ai/

Adds a vector database and Retrieval-Augmented Generation (RAG) service for semantic search and LLM-powered tooling.

Component Purpose
Weaviate Vector database for storing and querying AI embeddings
RAG Service FastAPI service providing semantic search over platform knowledge

When to add: You are building LLM-powered developer tooling, need semantic search over internal documentation, or want to experiment with AI-assisted platform capabilities.

Resource cost: 2–4 GB RAM, 2 vCPU additional.

the AI Extension (available in the repository under extensions/ai/)

Deployment Pattern

Extensions are self-contained ArgoCD Application manifests in the extensions/ directory. They are not included in the core platform/apps/ bootstrap, so they will not be deployed unless explicitly applied.

# Example: deploy the AI extension
kubectl apply -f extensions/ai/weaviate-application.yaml -n fawkes
kubectl apply -f extensions/ai/rag-service-application.yaml -n fawkes

Extension Design Principles

  • Opt-in: Extensions are never deployed by make dev-up or the core bootstrap.
  • Self-contained: Each extension directory includes all manifests, Helm values, and documentation needed to deploy it independently.
  • Documented trade-offs: Every extension README lists resource costs, prerequisites, and operational considerations.

See Also