Source-to-Vault Audit

Purpose

This ledger tests whether the vault can teach an objective independently. It is not a count of notes, links, modules, or questions.

Do not interpret in-progress as exam-ready or self-contained. A row is only verified when source, concept, route, procedure, lab evidence, and the three practice modes have been inspected.

Authoritative baseline

CourseSourceCurrent source versionRe-audit trigger
AZ-900Microsoft study guideSkills measured as of 2026-07-20; page updated 2026-06-22Source effective date or change log changes.
AZ-104Microsoft study guideSkills measured as of 2026-04-17; page updated 2026-03-19Source effective date or change log changes.
AZ-204Microsoft study guideFinal skills measured as of 2026-01-14; exam retired 2026-07-31; courseware retired 2026-05-29Retirement record, source change, or product behavior changes.
AZ-400Microsoft study guideSkills measured as of 2026-07-27; page updated 2026-06-26Source effective date or change log changes.

Source sweep — 2026-08-22

The current official study guides were re-read against the detailed ledgers. The AZ-900 inventory contains 57 atomic study-guide rows; the legacy ledgers contain 82 AZ-104 bullets and 86 AZ-400 bullets; and the initial AZ-204 inventory adds 37 atomic study-guide rows. The AZ-900 and AZ-204 rows are intentionally missing until actual targets are built and inspected.

This confirms coverage of the terms and operations named by the two study guides. It does not claim coverage of every Azure, Entra, GitHub, or Azure DevOps term that might appear in product documentation, because Microsoft states that related topics can also be assessed. Product-documentation changes remain re-audit triggers.

Audit rule

Use the project’s source-to-vault-audit skill and its evidence schema. In particular, “a glossary page exists” and “a question exists” are not proof that the learner can configure or diagnose the source objective.

Structural check on 2026-08-22: bash scripts/verify-learning-content.sh passed with 77 module hands-on sections, 50 standalone labs that each expose practical stages, three-mode scenario practice, source links, and a re-audit trigger, and 5,079 unambiguous active Obsidian wikilinks. This prevents a broken graph edge or an incomplete lab structure; it does not upgrade an objective’s pedagogical grade.

bash scripts/verify-source-audit-status.sh also checks that the detailed ledgers and this dashboard report the same grade totals. It checks accounting, not source accuracy or lab execution.

Phase 1 finding

The earlier coverage matrix was a structural mapping, not a source-to-vault verification. It treated linked concepts and a generic module lab as sufficient evidence. This audit replaces that interpretation.

The vault’s conceptual audit is complete for the current source baseline, but it is not a claim of live operational or exam readiness. The detailed ledgers currently contain 0 missing, 0 thin, 0 covered, and 270 verified conceptual rows. Practical status remains separate and visible in each ledger.

The separate course syllabus and prerequisite ledger adds eight covered teaching-baseline rows across the four courses; final verification still requires the cold-reader pass.

The two high-risk tables below preserve the initial Phase 1 snapshot. Their thin labels are historical; use the detailed ledgers for current row grades and the remediation queue for current work.

AZ-104 high-risk objective audit

Detailed, bullet-level ledger: AZ-104 Identity and Governance.

Detailed, bullet-level ledger: AZ-104 Storage.

Detailed, bullet-level ledger: AZ-104 Compute.

Detailed, bullet-level ledger: AZ-104 Networking.

Detailed, bullet-level ledger: AZ-104 Monitor and Recovery.

Detailed, bullet-level ledger: AZ-400 Build and Release Pipelines.

Detailed, bullet-level ledger: AZ-400 Security and Instrumentation.

Detailed, bullet-level ledger: AZ-400 Processes and Communications.

Detailed, bullet-level ledger: AZ-400 Source Control.

Detailed, bullet-level ledger: Retired AZ-204 application-development foundation.

Source objective (paraphrased)Inspected vault evidenceGradeExact gap / next evidence
Manage Entra users, groups, licenses, external users, and SSPRuser-account, groups, license-management, identity modules/questionsthinIndividually verify create/edit/query/guest-removal procedures in portal and CLI/Graph where relevant; split one generic identity lab into observable tasks.
Manage built-in Azure roles, scope assignments, and interpret accessrbac, scope, role-assignment, identity scenarioscoveredAdd a deterministic effective-access troubleshooting procedure using the portal and CLI; verify deny-assignment treatment from current docs.
Policy, locks, tags, subscriptions, budgets, Advisor, management groupsGovernance modules and scenariosthinVerify each portal operation and inheritance/failure behavior; separate budget alert and Advisor recommendation labs.
Storage firewalls/VNets, SAS, stored access policies, keys, Azure Files identity accessStorage notes/questionsthinThe concepts exist, but the audit has not confirmed a complete end-to-end Azure Files identity configuration and failure diagnosis.
Storage account, redundancy, object replication, encryption, Storage Explorer, AzCopyStorage modules and conceptsthinReproduce source prerequisites and cleanup for each operation; pricing/limit claims require product-doc citations.
File shares and Blob: tiers, soft delete, snapshots, lifecycle, versioningStorage notes/questionsthinCreate a distinct recovery lab with expected versions/snapshots, deletion, restore, and cleanup evidence.
ARM/Bicep interpretation, modification, deployment, export/conversionarm, bicep, az104-arm-bicep-change-labcoveredExecute the source-linked export/decompile, validation, what-if, deployment, failure, and cleanup steps before verified.
VM create/configure, encryption-at-host, moves, sizing/disks, zones/sets/VMSSVM notes/questionsthinAdd a source-checked procedure per move type and an observable disk/VMSS scale exercise.
ACR, ACI, Container Apps, sizing and scalingacr, aci, azure-app-container-apps, az104-container-sizing-scaling-labthinThe ACI/Container Apps sizing and scaling operation is now covered; add ACR and Container Apps provisioning evidence before this aggregate row can advance.
App Service plan/scaling/app/certs/custom DNS/backup/networking/slotsApp Service notes/questions, az104-app-service-backup-restore-labthinBackup/restore is now covered; one end-to-end lab still needs TLS, custom domain, private networking, and slot evidence.
VNet/subnet/peering/public IP/UDR/connectivity troubleshootingNetworking concepts and scenariosthinInspect actual routes/effective paths; add a public-IP and UDR procedure with a controlled broken route.
NSG/ASG/effective rules, Bastion, service endpoints, private endpointsnsg, asg, azure-bastion, service-endpoint, private-endpointcoveredVerify the distinct service-endpoint/private-endpoint lab and DNS failure evidence.
DNS, public/internal load balancers, load-balancing troubleshootingDNS/load-balancing concepts and scenariosthinAdd a health-probe failure procedure and internal-vs-public load balancer selection exercise.
Monitor metrics/log settings/KQL/alerts/Insights/Network WatcherMonitor notes and scenariosthinConfirm ingestion settings, query outputs, and alert-processing behavior in an executable lab.
Vaults, policies, backup/restore, Site Recovery failover, reports/alertsBackup concepts and labthinSeparately verify Recovery Services vs Backup vault support, restore evidence, test failover, and cleanup.

AZ-400 high-risk objective audit

Source objective (paraphrased)Inspected vault evidenceGradeExact gap / next evidence
GitHub Flow, feedback, Issues/Projects/Boards, traceabilityProcess path and scenariosthinProve an issue → PR → commit → build → deployment → incident feedback chain in both GitHub and Azure DevOps.
Dashboards, metrics, and queries for planning through operationsdevops-flow-metrics, instrumentation contentthinAdd real query/dashboard artifacts and expected interpretation, not only definitions.
Markdown/Mermaid, release/API documentation, Git-history automation, webhooks, Teams integrationDocumentation/communication modulesthinVerify real GitHub/Azure DevOps/Teams integration configuration; do not treat a Markdown note as integration evidence.
Branch strategies, PR workflow/protection/restrictionsGit notes/questionsthinAdd a branch-policy lab for GitHub and Azure Repos and a broken-policy diagnosis.
Large files, repo scaling, permissions/tags, Git recovery/removalgit-repository-scaling, Git scenariosthinAdd safe sandbox Git exercises for LFS, recovery, and removal; source-check destructive history operations.
Packages, feeds/views, dependency/artifact versioningPackage notes and pipeline scenariosthinProve publish/consume/retention flow in both a GitHub and Azure Artifacts context.
Pipeline tests, test agents/results, coverage, quality/release gatesCI content and labthinInspect an actual YAML implementation and gate outcomes; demonstrate a failed test and blocked release.
Runners/agents, triggers, YAML, execution order, templates, variables/groups, environments/approvalsPipeline path and 82 scenariosthinThis is well represented conceptually, but the audit still needs source-to-YAML evidence for every official sub-bullet.
Strategies, feature flags, ordered dependencies, downtime, hotfix, resilience, containers/binaries/scripts/database tasksRelease contentthinAdd a coherent deployment lab that includes a database-safe change and measured rollback decision.
IaC, desired state, Azure Machine Configuration, Deployment EnvironmentsIaC modules/notesthinVerify current product state and one working desired-state/on-demand-environment procedure.
Pipeline health/cost/concurrency/retention/classic-to-YAML migrationMaintenance notes/questionsthinAdd measurable health dashboard, retention test, and classic-migration artifact.
Service principals/managed identities, GitHub/Azure DevOps auth, service connections, security groups, access levels/projects/teamsExisting identity notes; new service-account, service-connection, security-groupthinThe canonical second-brain nodes were missing before this audit. Add both-platform configuration and permission-boundary exercises.
Key Vault, secretless auth, secure files, leak preventionSecurity notes/labthinSeparate Azure DevOps secure-files handling from Key Vault and OIDC exercises; test a deliberate authorization failure.
Defender/GHAS/CodeQL/container/dependency scanningSecurity notes/questionsthinVerify current product availability and configure an actual scan, finding, gate, remediation, and exception lifecycle.
Azure Monitor/Logs, Insights, GitHub insights/charts, pipeline alerts, telemetry/KQL/distributed tracingInstrumentation notes/labthinAdd actual query, trace, chart, and alert outputs for both platform and Azure telemetry paths.

Vocabulary and graph findings

FindingImpactRemediation state
Service connection was only canonically named subscription-connection; other pages referenced service-connection.A learner cannot tell pipeline authorization from Azure networking.Added service-connection with a direct contrast to service-endpoint and private-endpoint.
“Service account” had no canonical node.Human, service-principal, managed-identity, and OIDC roles could blur.Added service-account.
“Workload” was used as prose but not a graph node.The vault remains product-first rather than system-first.Added workload as a cross-domain integration node.
“Security group” has three incompatible meanings.Learners can apply identity, DevOps, or networking controls in the wrong plane.Added security-group disambiguation.
Azure DevOps security groups, access levels, secure files, and GitHub Project insights were only mentioned in surrounding prose.Crucial AZ-400 platform boundaries lacked canonical definitions and comparison links.Added azure-devops-security-groups, azure-devops-access-levels, azure-pipelines-secure-files, and github-project-insights. Their operational rows remain thin until labs exist.

Prioritized remediation queue

  1. Execute each source-linked training procedure and retain its requested success, failure, repair, and cleanup evidence.
  2. Re-read the cited product source and inspect rendered links before changing an individual covered row to verified.
  3. Build an evidence record for configuration, diagnosis, and distinction practice per objective; questions and notes alone do not verify it.
  4. Re-audit immediately when a study guide, product behavior, permission model, or service state changes.

Structural coverage matrix · Readiness standard · workload · service-connection · security-group