Skip to content

Validate & Audit

Validation checks three things:

  1. JSON schema compliance — structure, types, required fields, enum values
  2. Referential integrity — every ID reference resolves to an existing entity
  3. Circular reference detection — no infinite parentId chains

Ask your agent:

“Validate the workspace. If there are errors, fix them and validate again.”

The agent handles the fix-validate cycle on its own. You only see the final result.

The audit runs 70+ deterministic checks on your architecture data. Not AI opinions — rules-based assessments.

“Run a full audit on the workspace”

“Show me only critical compliance findings”

“Fix all data quality warnings automatically”

The audit is grouped into categories that reflect the kinds of question an architect asks of a landscape. Broadly:

  • Structural integrity — duplicate names, self-references, hierarchy depth, invalid references.
  • Data quality — missing classifications, fit scores, criticality, descriptions.
  • Architecture consistency — hosting / protocol mismatches, cross-mapping consistency, stack cohesion.
  • Lifecycle coherence — date ordering, end-of-life dependencies, phase-out state.
  • Network analysis — integration hubs, circular dependencies, isolated applications.
  • Portfolio health — TIME distribution, critical-app risks, capability coverage.
  • Compliance and risk — GDPR / PCI / SOX correlation, auth security, data retention.
  • Technology risk — end-of-life components, vendor concentration, monoculture.
  • Initiative alignment — unlinked initiatives, conflicting timelines, migration planning.
  • Organizational coverage — missing ownership, orphaned entities.
  • Data governance — unclassified data, PII without controls, source-of-record conflicts.
  • Strategic alignment — gaps between declared strategy and what the portfolio actually does.
  • Redundancy and complexity — duplicate capabilities, integration sprawl.

Each finding has a severity level: critical, warning, or info. The canonical list of individual checks lives in the analyzer code; expect the set to grow over time as new checks are added.

After editing, validating, and auditing:

“Push the workspace. Name the change request ‘Q2 Architecture Cleanup’.”

The agent creates an Architecture Change Request (ACR). You review and approve it in Albumi’s UI.