CodeRadius LogoCodeRadius Docs
Explore

System Registry

The auto-generated service catalog — a live inventory of every repository, service, and team in your architecture graph.

System Registry

The System Registry is the default tab of the Architecture Dashboard. It is an auto-generated, queryable catalog of every repository, service, and team that CodeRadius has ingested — a live replacement for manually maintained Backstage catalogs, spreadsheet inventories, and tribal knowledge.

# Open the full dashboard (System Registry is the first tab)
cr ui

# Focus exclusively on the System Registry
cr ui --focus inventory

What It Shows

The System Registry is organized into three sub-tabs:

Repositories

A sortable, filterable table of every repository in the graph.

ColumnDescription
RepositoryFull org/group/repo path, clickable link to the source
BranchThe ingested branch (e.g., main, master)
DepthThe depth of analysis performed: CONTRACTS, SEMANTIC, or STRUCTURE
StackProgramming languages detected in the repository
TeamsTeams that own services stored in this repository
PulseLiveness indicator: commit velocity over the last 12 months

Analysis Depth

BadgeMeaningHow to Get It
CONTRACTSFull semantic extraction — intent, dependencies, capabilities, API call graph, data contracts (produced/consumed payload schemas)cr analyze code --depth contracts
SEMANTICSemantic code scan — intent, infrastructure dependencies, capabilities, API call graph via LLM analysis. Skips payload field extraction.cr analyze code
STRUCTUREStructural scan only — repository registered, agentic context and structural files detected, no code analysiscr analyze code --depth structure

Pulse (Liveness)

The Pulse column shows commit velocity over the last 12 months, with a tooltip displaying the exact commit count and unique author count:

LevelBadgeCommit Range
HighGreen, pulsing dot100+ commits
ActiveCyan, pulsing dot20–100 commits
LowYellow, static1–20 commits
DormantDim, static0 commits

Dormant repositories are still listed in the System Registry for completeness, but they are excluded from aggregated metrics in the Agentic Radar to prevent skewing maturity distributions.


Services

A table of every service extracted from the graph, with ownership and complexity metrics.

ColumnDescription
ServiceThe service name as detected from project structure or configuration
OwnerThe team that owns this service (from Backstage catalog or git org)
StackProgramming languages used by this service
RepositoryThe repository that contains this service, clickable
FunctionsTotal number of functions indexed by CodeRadius
APIsNumber of exposed API endpoints detected
DependenciesTotal package dependency count

Dim em-dash () values indicate missing data — for example, a service with no declared team owner or a repository with no detected API endpoints.


Teams

A table of every team in the graph with their aggregate ownership metrics.

ColumnDescription
TeamThe team name as registered in the graph
TypeTeam classification (e.g., Stream-aligned, Platform, Enabling)
ServicesNumber of services owned by this team
RepositoriesNumber of repositories owned by this team
StackAggregate programming languages across all owned services

Header Stats

When the System Registry is active, the header stats bar shows:

MetricDescription
RepositoriesTotal number of repositories in the graph
ServicesTotal number of services extracted
TeamsTotal number of teams detected
Deep ScanCount of repositories with full semantic extraction (with tooltip breakdown)

Filtering and Searching

Every table in the System Registry supports:

  • Global search — A search bar that filters across all visible text content and hidden searchValue fields (e.g., repository URLs, hashes)
  • Column filters — Clickable filter icons on supported columns that show a dropdown of unique values for that column
  • Sorting — Click any column header to sort ascending/descending
  • Export — The Export button downloads the currently filtered table as a CSV file

Relationship to cr sync

The System Registry is populated during the sync pipeline:

  1. cr analyze code --depth structure — Scans repositories for metadata (team ownership, agentic configs, structural files). Creates STRUCTURE-level entries.
  2. cr analyze code — Full LLM-powered semantic analysis: intent, infrastructure dependencies, capabilities, API call graph. Creates SEMANTIC-level entries.
  3. cr analyze code --depth contracts — Extends CODE with data contract extraction: produced and consumed payload schemas (field names and types) for every I/O boundary. Creates CONTRACTS-level entries.

Each subsequent analysis level builds on the previous one. A repository can be upgraded from STRUCTURESEMANTICCONTRACTS by re-running analysis with a higher depth.


Further Reading

On this page