AI diagram generator

AI GCP architecture diagram generator.

From the Terraform in your repository, with no project access and nothing uploaded.

Download for macOS
v0.1.33 · .dmg · Apple Silicon & Intel
Signed & notarized by Apple · opens without a Gatekeeper warning
sha256 698955a0187bc039f4c74f5d05a9f10fbb27376a45788a0a241d1326b73873c7
Download for Windows instead
$curl -fsSL https://lucidtrain.com/install.sh | sh

Sources it reads

Terraform with the google provider, Deployment Manager configurations, and Kubernetes manifests for GKE workloads. Terraform is the common case and reads well, since the reference structure gives reliable edges without inference.

How Lucid Train draws one

Describe what you want, or point it at a repository. The agent emits a semantic graph of nodes and edges with no coordinates, and the ELK layout engine computes positions and orthogonal routing. That separation is the reason the output cannot come back with boxes overlapping or arrows crossing through cards, which is the characteristic failure when a language model is asked to place things on a canvas itself. It also means editing is editing the graph: change a component and the layout re-solves rather than drifting out of alignment.

Cloud Run and Functions as application components

GCP systems are frequently a set of Cloud Run services connected by Pub/Sub, which is a shape that draws cleanly: services in the application tier, Pub/Sub topics as the edges between them rather than as boxes. That keeps the diagram close to how people describe these systems, where the topic is the connection and not a destination.

Projects as sections

Multi-project layouts are common in GCP, often for billing or isolation rather than for architectural reasons. Drawing projects as sections keeps that structure visible without letting it become the organising principle, since the tier a component belongs to usually says more about the system than which project it is billed to.

Local, offline, your models

It runs as a desktop application. Against a local Ollama model the whole thing works offline with no API key and no telemetry, which is the practical difference from every browser-based diagram tool: your architecture never leaves the machine.

GCP services by tier

ServiceTier
Cloud CDN, Cloud Load BalancingEdge
API Gateway, ApigeeEdge
Cloud Run, GKE, Compute EngineApplication
Cloud Functions, WorkflowsApplication
Cloud SQL, Spanner, FirestoreData
Cloud StorageData
MemorystoreData
Pub/Sub, Cloud TasksDrawn as edges
Identity Platform, Secret ManagerAuthentication boundary
Cloud Monitoring, Cloud TraceInfrastructure

Try it

shell
$ Read the Terraform here and draw the GCP architecture. Group by tier, show Pub/Sub topics as edges, and mark project boundaries.
FAQ

Questions

Does it need GCP credentials?

No. It reads files on disk.

Does it handle multi-project setups?

Yes, with projects drawn as sections. Cross-project edges are worth marking because they are where IAM complexity usually lives.

Can it diagram GKE workloads?

Yes, from the manifests, and it is usually clearer as a separate diagram from the infrastructure that creates the cluster.

Does it use official Google Cloud icons?

Bundled service icons that render offline, with a generic glyph where a specific mark is not bundled.

Related