Changelog

• September 30, 2026: Spread objective monitoring across four stable monthly cohorts on the 1st, 8th, 15th, and 22nd. Added manual single-exam runs, candidate diffs, and a checklist issue for detected changes so guides can be reviewed and updated one exam at a time. Kept the combined full-library maintenance sweep manual; scheduled source-health and catalog discovery checks continue to report without changing guides.

• September 30, 2026: Reconciled the roadmap's published-guide count with the catalog: 223 guides across 26 vendors, including six GitHub guides. Clarified that the 222-guide denominator in the older rubric-1 audit summary is historical.

• September 30, 2026: Made the AI-generated, best-effort status of the AB-100 lab explicit on the lab page and starter README. Clarified that 13 synthetic local tests passed, while the proposed service workflow has not been tested in a real tenant or independently reviewed; added the same boundary to the lab index.

• September 30, 2026: Reworked the certification lifecycle page into source-linked dated events and a vendor research queue. Added exact English skills dates from 41 Microsoft Learn study guides (40 new dated skills records) and the AIB-C01 beta delivery date, raising entries with a dated event from 8 to 46 of 226. The other 180 remain explicit research gaps; month-only headings were not assigned invented days. The page is a generated site snapshot, not a live vendor feed.

• September 30, 2026: SPLK-5002 deep review: Mapped 19 detailed objectives and reviewed 40 answers; added a synthetic detection worksheet and qualified ES findings, SOAR pairing, and playbook recovery by release. Documented conflicting ES 8.5 output guidance, twelve unexecuted product labs and source-access limits.

• September 30, 2026: SPLUNK-ADVANCED-POWER-USER deep review: Mapped104 objectives; corrected blueprint/course arithmetic and tstats/token/transaction assumptions. Documented ten unexecuted labs and source/training follow-ups.

• September 30, 2026: SPLUNK-CORE-POWER-USER deep review: Mapped 32 objectives; clarified null-field scope, Boolean precedence, transaction boundaries, alias collisions and calculated-field dependencies. Recorded paper examples, eight unexecuted labs and source/training follow-ups.

• September 30, 2026: SPLUNK-CORE-USER deep review: Mapped 40 objectives; clarified command order, missing keys and sort limits with original paper predictions. Qualified historical training and unverified book metadata; documented all eight unexecuted labs and next-pass fixes.

• September 30, 2026: Added source-backed certification lifecycle tracking and a fourth aliases column in the LMS query export. AB-100 retains its former name alongside the current Expert credential name. The dated page distinguishes effective updates, announced future updates, release stages and inventory checks; unconfirmed dates remain explicit gaps.

• September 30, 2026: PCEI-30-01 deep review: Mapped33objectives separately from36items and expanded20answers; exact Python/NumPy workbook passes21methods and20observations per mode. Added training-only scaling,metric/axis/gradient evidence; recorded practice-release conflicts and unexecuted optional frameworks,plots,model trials andhumanreview.

• September 30, 2026: PCEA-30-01 deep review: Mapped 46 objectives and expanded 18 answers; three exact original files pass 25 methods and 23 observations per mode. Added owned HTTP, strict data, idempotent replacement and byte-recovery evidence; preserved beta status, repaired the book link, and recorded remaining scheduler/notification/full-lab and human-review limits.

• September 30, 2026: PCES-30-01 deep review: Mapped 45 detailed objectives; three exact original files execute 30 methods and 21 observations per mode, including owned loopback, context encoding, pseudonymized reports, PDF metadata and byte-restore trust limits. Removed the wrong book recommendation; missing tools, broader labs and human review remain explicit.

• September 30, 2026: PCAT-31-01 deep review: Mapped 34 objectives; exact 25-method/31-probe multi-module workbook in both Python modes; actual fixture/marker/order, coverage-boundary and red–green–refactor evidence; corrected learning entitlements and retained publisher, integration and human-review limits.

• September 30, 2026: PCET-30-01 deep review: Mapped 22 objectives; exact original 25-method/55-probe workbook in normal and optimized modes; measured coverage/mutant and pdb evidence; corrected course entitlements and double guidance; preserved named-tool, publisher and human-review blockers.

• September 29, 2026: PCAD-31-02 deep review: Mapped 48 objectives; added an original 86-check NumPy/SciPy/SQLite/local HTTP workbook and a prepared-only Pandas 3/plotting companion; corrected learning and product claims, with missing runtimes and publisher/human-review questions recorded.

• September 29, 2026: PCED-30-02 deep review: Mapped40objectives; original112-checkPython/NumPyworkbook, reconciledCSVledger andexportedchart; correctedlearningduration/access andpractice-product limits; broaderlab andhumanreview remainpending.

• September 29, 2026: PCPP-32-101 deep review: Mapped 24 objectives, expanded 23 answers and added five original Python files. Eight normal/optimized runs passed173 checks per mode, including a withdrawn Tk desk and disposable HTTP/TCP fixtures. Corrected geometry, transaction, persistence and course-duration boundaries; policy/access and human-review limits remain explicit.

• September 29, 2026: CPP-22-02 deep review: Mapped 34 objectives and expanded 30 answers; seven original C++17 files passed 16 hosted workbook runs, 16 diagnostic probes and four legacy mode probes. Provider arithmetic, eligibility/course/policy conflicts and broader lab/human review limits remain explicit.

• September 29, 2026: CPA-21-02 deep review: Mapped 35 objectives and expanded 30 answers; six original C++17 files passed 13 hosted workbook runs, 16 diagnostic probes and ten mode probes. Source/booking conflicts, broader labs and human review remain explicit.

• September 29, 2026: CLP-12-01 deep review: Mapped 29 objectives and expanded 30 answers; five complete original C11 programs passed 17 hosted workbook runs plus seven dialect/header checks. ThreadSanitizer crashes, booking/source conflicts and human review remain explicit.

• September 29, 2026: CLA-11-03 deep review: Mapped 23 objectives and expanded 30 answers; fixed allocation, overflow, lifetime and input contracts. Five C files executed remotely with 181 loader and 53 language checks on both compilers, selected sanitizer/optimization runs and six negative diagnostic checks. Resource/policy limits and human review pending.

• September 29, 2026: WDA-41-01 deep review: Mapped 40 objectives and expanded 40 answers. Fifteen exact public blocks, 14 original assets, 42 CSS and 90 browser checks, five Nu-clean pages, two CSS3 passes and actual browser Sass/Bootstrap evidence. Ten wider labs proposed; duration/access and human limits retained.

• September 29, 2026: WDE-40-01 deep review: Mapped 40 objectives and improved 40 answers. Seven exact public blocks, 54 core and 128 browser checks, three Nu-validated HTML pages and eleven reproducible assets. Delivery, access and human-review limits remain explicit; eight broader activities proposed.

• September 29, 2026: JSA-41-01 deep review: Mapped 40 objectives and improved 40 answers. Seven exact public blocks verified: 143 core and 84 scenario checks in Node/Chrome plus 36 browser API/UI checks. Descriptor, sparse-array, validation and promise contracts corrected; channel/access limits and eight proposed broader activities remain explicit.

• September 29, 2026: PCAP-31-03 deep review: Mapped 21 Python objectives, improved 27 answers and completed three runnable scenarios. Twelve public blocks passed 78 core and 98 additional checks on CPython 3.13.14. Provider duration and import/sampling/exception/string/I-O contracts corrected; policy/access conflicts and human review remain explicit.

• September 29, 2026: CLE-10-01 deep review: Mapped all 36 C objectives, improved 30 answers and completed three worked programs. Eight public blocks executed remotely; 72 core and 118 scenario checks passed on both compilers plus nine sum boundaries. Resource/access and reference-quality limits documented; human review pending.

• September 29, 2026: JSE-40-01 deep review: Mapped all 30 unweighted scope bullets, improved 40 answers and completed three worked programs. Eight public blocks executed; 114 core and 109 scenario checks plus actual dialogs/timers/debugger evidence passed. Resource metadata and access/policy limits documented; human review pending.

• September 29, 2026: CPE-20-01 deep review: Mapped all 26 C++ objectives, repaired input/pointer/string explanations and completed three worked programs. Ten public blocks executed remotely; 67 workbook and 60 scenario checks passed on two compilers. Resource access and vendor wording limits documented; human review pending.

• September 29, 2026: PCEP-30-02 deep review: Mapped 84 syllabus bullets and corrected Python tracing explanations; four public code blocks passed 136 offline checks with three worked programs. Refreshed catalogs and documented current-version, source-access and vendor-page discrepancies; human review pending.

• September 29, 2026: CRISC deep review: Mapped 68 canonical entries; corrected QAE count and added 40 answers, three worked scenarios and 52 executed schema, simulation and portfolio checks.

• September 29, 2026: CISM deep review: Mapped 72 current entries; corrected November weights/architecture changes and added 40 answers, three worked scenarios and 47 executed management checks.

• September 29, 2026: CISA deep review: Mapped 60 subtopics and 43 tasks; added 40 answers and 47 executed checks including SQLite backup/restore, and updated ITAF, QAE and 2027 CPE guidance.

• September 29, 2026: SPLK-5001 deep review: Mapped 24 objectives; added 40 answers and 49 executed checks, with corrections for renewal, summary completeness, entity identity and version-specific detection behavior.

• September 29, 2026: NSE-4-FORTIOS deep review: Recorded the undated 8.0 successor while preserving the available 7.6 scope and history; added 40 answers, 41 executed checks and corrections for policy, NAT, inspection, HA, routing and IPsec.

• September 29, 2026: PANW-CYBERSECURITY-APPRENTICE deep review: Checked the full current blueprint and September handbook/FAQ, added 40 answered prompts and 43 executed Python checks, and clarified network, telemetry, identity and learning-resource boundaries.

• September 29, 2026: SERVICENOW-CAD deep review: Aligned all 22 canonical subtopics and weights with archived history; added 40 answered prompts, 38 executed JavaScript checks and API, security, integration and release evidence.

• September 29, 2026: SERVICENOW-CSA deep review: Restored 30 separate canonical subtopics with archived history, added 40 answered prompts and 34 executed Python/SQLite checks, corrected course/catalog links and qualified ACL, import, logic and transport evidence.

• September 29, 2026: MONGODB-ASSOCIATE-ATLAS-ADMINISTRATOR deep review: Added 40 answered prompts and 36 executed Python/SQLite checks, corrected restore-command guidance, clarified access/recovery/edition boundaries and catalog timing, and preserved the blocked detailed objective boundary.

• September 29, 2026: MONGODB-ASSOCIATE-DATA-MODELER deep review: Added 40 answered prompts and 34 executed model/migration predictions; clarified validation, reconciliation and rollback, recorded conflicting exam/catalog measures, and retained the blocked current objective boundary.

• September 29, 2026: MONGODB-ASSOCIATE-DEVELOPER deep review: Added 40 answered prompts and 38 executed prediction checks, clarified query/atomicity/index/driver contracts and catalog durations, and preserved the blocked current objective-body boundary.

• September 29, 2026: SALESFORCE-AGENTFORCE-SPECIALIST deep review: Corrected26objectivewording with historical archives,added40answers and35optional executedchecks,clarified masking/identity/testing boundaries andupdated learning metadata.

• September 29, 2026: SALESFORCE-PLATFORM-DEVELOPER deep review: Compared 21 objectives, added 40 answers and 31 executed bulk/transaction checks, clarified API67 security and CSP rules, and recorded source/catalog limitations.

• September 29, 2026: SALESFORCE-PLATFORM-APP-BUILDER deep review: 26 objectives manually compared,40 answered prompts and28 executed local relationship/transaction/dependency checks; derived-field access,approval locking,deployment evidence andmaintenance discrepancy documented.

• September 29, 2026: SALESFORCE-PLATFORM-ADMINISTRATOR deep review: eight unchanged domains,40 answered prompts and30 actual local CSV/SQLite/access checks; Flow context/rollback and agent scoping clarified, with source and training limitations recorded.

• September 29, 2026: NCP-AIO deep review: Mapped 31 canonical topics, recorded a conflicting official PDF and added 25 local rendering/accounting checks with 48 answered prompts.

• September 29, 2026: NCA-AIIO deep review: Mapped 22 unchanged objectives, added 34 executed telemetry/capacity checks and answered all 40 readiness prompts.

• September 29, 2026: NCA-GENL deep review: Mapped 33 current topics, repaired the page monitor and added 23 executed NumPy checks with 40 answered prompts.

• September 29, 2026: CISSP deep review: Mapped 62 objectives, updated waiver and risk guidance, and added 23 executed local checks with 64 answered prompts.

• September 29, 2026: CCSP deep review: Mapped38 current objectives, reconciled FIPS/CSA transitions and catalogs, and added24 executed encryption checks with48 answers.

• September 29, 2026: SSCP deep review: Mapped36 current objectives, reconciled PDF/course inconsistencies, and added11 executed TLS cases with48 original answers.

• September 29, 2026: CC deep review: Mapped 19 current objectives, reconciled program and policy changes, and added 34 executed control/recovery checks with 48 answers.

• September 29, 2026: GES-C02 deep review: Corrected search/agent access and document API contracts; refreshed training evidence with 43 executed local checks and 48 answers.

• September 29, 2026: DEA-C02 deep review: Updated streaming, task and replication contracts; refreshed catalog evidence and added 40 executed CDC checks with 48 answers.

• September 29, 2026: COF-C03 deep review: Updated authentication, pipeline and performance reasoning; verified training metadata and added 24 executed local checks with 48 answers.

• September 29, 2026: SOL-C01 deep review: Preserved retired scope, verified the replacement course, updated notebook/constraint/load/role/Cortex guidance and added 45 executed local checks with 42 answers.

• September 29, 2026: 200-901 deep review: Mapped 61 objectives, fixed Boolean parsing and incomplete API collection, added 67 executed checks and 48 answers, clarified protocol/tool limits and refreshed learning evidence.

• September 29, 2026: 200-301 deep review: Mapped 53 current and 29 future objectives separately, corrected DHCP transition and routing/LACP/ACL teaching, added 48 executed checks and 48 answers, and refreshed catalog evidence.

• September 29, 2026: 100-160 deep review: Mapped 23 detailed objectives, reconciled NIST incident guidance and CCST renewal, added 40 executed evidence checks and 48 answers, and refreshed learning catalogs.

• September 29, 2026: 100-150 deep review: Mapped the actual25-objective PDF, corrected dated renewal policy, added40executed packet/subnet checks and48answers, and qualified current catalog evidence.

• September 29, 2026: GOOGLE-PROFESSIONAL-AGENTIC-ARCHITECT deep review: 31 mapped considerations plus28 tool entries,31 executed local action-control checks,52 answers,eight proposed labs and updated beta/catalog evidence.

• September 29, 2026: GOOGLE-PROFESSIONAL-MACHINE-LEARNING-ENGINEER deep review: 52 mapped considerations,36 executed local training/evaluation checks,48 answers,eight proposed cloud labs and updated source/catalog evidence.

• September 29, 2026: GOOGLE-PROFESSIONAL-CLOUD-SECURITY-ENGINEER deep review: 70 mapped considerations, 35 executed local envelope-encryption checks, 48 answers, eight proposed cloud labs and updated source/catalog evidence.

• September 29, 2026: GOOGLE-PROFESSIONAL-DATA-ENGINEER deep review: 67 mapped considerations, 40 executed local SQL/window checks, 48 answers, eight proposed cloud labs and refreshed course/catalog evidence.

• September 29, 2026: GOOGLE-PROFESSIONAL-CLOUD-ARCHITECT deep review: 101 mapped statements, four case analyses, 32 executed local architecture checks, 48 answers, eight proposed labs and current learning-resource evidence.

• September 29, 2026: GOOGLE-ASSOCIATE-CLOUD-ENGINEER deep review: 94 mapped considerations,32 executed local operations checks,50 answers,eight proposed cloud labs,explicit lifecycle-baseline initialization and current resource comparison.

• September 29, 2026: GOOGLE-GENERATIVE-AI-LEADER deep review: 61 published considerations, 27 executed local checks, stronger data/identity/tool boundaries, 48 answers, eight proposed labs and current learning-resource evidence.

• September 29, 2026: GOOGLE-CLOUD-DIGITAL-LEADER deep review: 69 detailed considerations, 26 executed local checks, app/platform and billing-control corrections, 44 answers, eight proposed decision labs and refreshed public learning evidence.

• September 29, 2026: CKS deep review: 26 competencies; actual PDF reconciliation; 17 real signature checks and 12 audit-model checks; 46 answers and eight proposed cluster labs; refreshed public evidence.

• September 29, 2026: CKAD deep review: 24 competencies, 27 executed Helm checks, native sidecar and rollout corrections, 46 answers, eight proposed cluster labs, current catalog and dated tooling support.

• September 29, 2026: CKA deep review: 27competencies;22actual kubectl/Kustomize checks;EndpointSlice/PDB corrections;Gateway/storage/recovery distinctions;46answers and eight proposed cluster labs;current catalog and retainedv1.35watch.

• September 29, 2026: LFCS deep review: 34 competencies mapped; original OpenSSL/TLS exercises passed 38 checks and six loopback handshakes; service/resize/identity boundaries clarified, 46 answers and eight proposed labs strengthened; native Linux blocker deferred.

• September 29, 2026: LFCA deep review: 22 competencies mapped; original Git/Bash exercises passed 35 checks; package trust, service state, permissions and project/licensing teaching strengthened; 46 answers and eight proposed labs; native Linux blocker deferred.

• September 29, 2026: CV0-004 deep review: 33 numbered objectives mapped; experience, community/AI/IoT and state/retry boundaries strengthened; original Terraform/SQLite examples passed 61 local checks; 46 answers and eight proposed labs expanded.

• September 29, 2026: XK0-006 deep review: 29 numbered objectives mapped; umask/ACL, current tools and orchestration/monitoring teaching improved; 44 local MSYS/Python checks passed; native Linux runtime blocker deferred and baseline history retained.

• September 29, 2026: SY0-701 deep review: 28 numbered objectives mapped; original crypto/certificate examples passed 34 checks, current identity/incident/risk distinctions added, 46 answers strengthened; duration-only metadata accepted with history retained.

• September 29, 2026: N10-009 deep review: 25 numbered objectives mapped; routing/DNS/MTU and modern-network teaching strengthened, 72 local checks and 24 loopback DNS exchanges passed; duration-only baseline change archived and accepted.

• September 28, 2026: 220-1202 deep review: 36 numbered objectives mapped; cloud productivity, AI and recovery guidance expanded, 42 answered checks strengthened and 18 native copy/restore checks passed.

• September 28, 2026: 220-1201 deep review: 27 numbered objectives mapped; methodology scope clarified, networking/hardware examples expanded, 34 local checks passed and course-catalog discrepancies disclosed.

• September 28, 2026: FC0-U71 deep review: 31 numbered objectives mapped; internet/wireless gaps repaired, three original examples and 30 local checks, current password guidance and 42 answers added.

• September 28, 2026: EX267 deep review: 43 main-page tasks mapped; PDF/serving terminology differences retained, 39 local KFP checks and versioned implementation/support boundaries added.

• September 28, 2026: EX378 deep review: 64 tasks/PDF compared; monitor cutoff repaired, 40 answers and 18 actual local Quarkus tests added with full-service/restart limits explicit.

• September 28, 2026: EX294 deep review: 56 tasks and the V26K PDF compared; 40 answers, typed template/candidate validation and 41 actual local checks with Ansible/RHEL execution limits explicit.

• September 28, 2026: EX280 deep review: 55 tasks and the V422 PDF compared; 40 answers, 24 local rendering/selector checks, SCC/OLM/TLS boundaries and a retained exam-version blocker.

• September 28, 2026: EX200 deep review: map 62 RHEL 10 tasks; answer all 40 checks; add persistence and system-state evidence; execute 21 local Bash/INI/arithmetic checks with VM limits explicit.

• September 28, 2026: DATABRICKS-ASSOCIATE-DEVELOPER-APACHE-SPARK deep review: map 32 objectives; repair JDBC, cast, watermark and AQE guidance; add 12 answers; record Spark/WSL execution blocker and 10 portable SQL checks.

• September 28, 2026: DATABRICKS-GENERATIVE-AI-ENGINEER-ASSOCIATE deep review: map 56 objectives; repair retrieval authorization and scorer coverage guidance; execute 27 local retrieval/PyFunc/MLflow checks.

• September 28, 2026: DATABRICKS-MACHINE-LEARNING-PROFESSIONAL deep review: map 47 objectives; track profiling-window ambiguity; repair monitoring and online-store references; execute 20 local PyFunc/Optuna/MLflow checks.

• September 28, 2026: DATABRICKS-MACHINE-LEARNING-ASSOCIATE deep review: map 48 objectives; clarify AutoML/Hyperopt support, feature contracts and serving releases; execute local ML exercise with 22 checks.

• September 28, 2026: DATABRICKS-DATA-ENGINEER-PROFESSIONAL deep review: map 45 detailed objectives; add serverless retry controls, CDC/recovery/purge boundaries, corrected aggregates and 13 bounded local assertions.

• September 28, 2026: DATABRICKS-DATA-ENGINEER-ASSOCIATE deep review: map 33 objectives; track DENY scope mismatch; repair schema recovery, deterministic survivorship, union and job-branch guidance; add conversion coverage and bounded model checks.

• September 28, 2026: DATABRICKS-DATA-ANALYST-ASSOCIATE deep review: map 39 detailed objectives; distinguish Genie product terminology and access identities; repair identifier/null examples; add ten answers and verify 18 portable SQL assertions.

• September 28, 2026: VAULT-OPERATIONS-PROFESSIONAL deep review: confirm the 1.16 baseline/rename; strengthen audit, health, seal, Sentinel and Agent guidance; add worked decisions and 20 answers; verify local ACL/audit behavior.

• September 28, 2026: VAULT-ASSOCIATE-003 deep review: map 40 objectives, flag conflicting official Vault versions, clarify policy/KV/token/client controls and add five worked decisions plus 20 answers.

• September 28, 2026: TERRAFORM-AUTHORING-OPERATIONS-PROFESSIONAL deep review: map 27 objectives, restore the Terraform 1.6 boundary, clarify state/testing/governance and Azure availability, add 20 answers, and validate local module migration plus three native tests.

• September 28, 2026: TERRAFORM-ASSOCIATE-004 deep review: map 37 objectives, clarify secret/state/governance controls, repair lab sequencing, add 20 answers and four decisions, and execute two local Terraform 1.12.2 labs.

• September 28, 2026: SCS-C03 deep review: map 70 skills, add three-language retirement, repair scope link, clarify IAM/KMS/detection, answer 42 checks and add five worked decisions.

• September 28, 2026: AIP-C01 deep review: map 98 skills, clarify AgentCore/RAG/evaluation and lifecycle boundaries, add five worked development decisions and bounded CI reading.

• September 28, 2026: DOP-C02 deep review: map 140 bullets, update native ECS deployments and service lifecycle, add five worked operating decisions; retain App Runner date conflict.

• September 28, 2026: SAA-C03 deep review: map 189 bullets, record Italian delivery retirement, clarify consistency/storage/scaling, repair order publication and add five worked decisions.

• September 28, 2026: DEA-C01 deep review: map 120 skills, add 40 answers and five worked cases, fill pipeline engineering gaps, and update Redshift/CloudTrail Lake and managed-table guidance.

• September 28, 2026: SOA-C03 deep review: map 53 skills, add AI monitoring and storage synchronization guidance, explain tool/action boundaries, and add four worked checks and five answers.

• September 28, 2026: AIF-C01 deep review: correct revision 1.1/date, map 69 objectives, add five worked cases and 35 answers, explain agent controls, and record availability/source-list limits.

• September 28, 2026: CLF-C02 deep review: map 135 detailed bullets, add 30 answers and four worked examples, update support-plan retirement and new signup/cost boundaries, and fill service-category gaps.

• September 28, 2026: ANS-C01 deep review: correct the passing score, map 177 detailed bullets, add four packet-path exercises, improve routing/multicast/controller coverage and flag App Mesh support end; retain vendor conflicts.

• September 28, 2026: SAP-C02 deep review: map 187 knowledge/skill bullets, add four worked cases, qualify migration/platform availability and recovery articles, and preserve unresolved vendor dates.

• September 28, 2026: review all 101 DVA-C02 skills, repair order-event reliability, distinguish partial retries, add 40 answers and four worked decisions, and flag X-Ray SDK maintenance. C02 delivery-date conflict remains.

• September 28, 2026: review all 58 AIB-C01 skills, add four worked business cases and 40 answered checks, flag historical CAF-AI guidance and assess practitioner articles. Beta-duration wording remains unresolved.

• September 28, 2026: review all 107 MLA-C02 skills, add five examples and 42 answers, refresh credential/service/catalog guidance and evaluate two useful AWS articles. GA-date disagreement remains open.

• September 28, 2026: review MLA-C01, map 70 skills from the historical PDF, repair reader links and qualify transition/service availability. Add four worked examples and 40 answers; missing current blueprint pages remain a blocker.

• September 28, 2026: reserve the first five pending certificates—GH-900, GH-200, GH-100, GH-500 and GH-600—for the weekly notification pilot, with manual baselines and five required notification scenarios. Conditional email delivery remains pending.

• September 28, 2026: add GH-300's runnable boundary/context workshop, seven requirement tests, verified threshold and boolean mutations, two prompts and a practitioner packet. Copilot session evidence is pending.

• September 28, 2026: add AI-103's runnable maintenance-assistant contracts: five retrieval fixtures, approval/asset/duplicate/timeout drills and 13 passing local tests. Azure and multimodal validation remain pending.

• September 28, 2026: add the runnable AB-100 release-gate workshop, 13 local checks, a deliberately blocked starter, and service/practitioner review packets. Local results do not establish live service behavior.

• September 28, 2026: recheck all 14 Microsoft certification blockers against 30 primary pages and document closure criteria. All remain open; future checkpoints and accepted blueprints are preserved. See the follow-up.

• September 28, 2026: review all 73 MO-310 tasks, add five examples, ten labs and 32 answered checks, and clarify media losses and current SmartArt accessibility. All 55 existing Microsoft/Office guides now have deep-review records; 14 source blockers remain. See the PowerPoint report.

• September 28, 2026: map all 48 MO-211 tasks and add six worked examples, ten labs and 32 answered checks covering advanced formulas, calculation freshness and Pivot aggregation. See the Excel Expert report.

• September 28, 2026: review all 65 MO-210 tasks, correct named-function scope and add six worked examples, ten labs and 32 answered checks with independently checked expected results. See the Excel Associate report.

• September 28, 2026: review MO-111 against all 47 PDF tasks, expanding templates, fields, macro boundaries and merge validation with five examples, nine labs and 32 answered checks. See the Word Expert report.

• September 28, 2026: review MO-110 against all 70 detailed PDF tasks, add practical Word guidance and include Microsoft Office in the recurring review queue. Refresh the Microsoft portfolio retirement/objective scan; existing notices and future revisions remain unchanged. See the MO-110 report.

• September 28, 2026: complete the SC-500 deep review across 87 objectives, with seven examples, ten labs and 48 answered checks. Update agent/AI, Key Vault, Sentinel and release boundaries plus learning catalogs. All 50 active Microsoft guides now have deep reviews; unresolved source blockers remain visible. See the SC-500 report.

• September 28, 2026: deeply review PL-900’s 37 objectives; add six examples, eight labs and 37 answered checks. Correct uploaded-file access, update app-building/policy/flow boundaries and compare learning catalogs. See the PL-900 report.

• September 28, 2026: deeply review MB-820’s 75 objectives; add six worked examples, ten labs and 48 answered checks. Correct try/commit, API batching and sandbox replica assumptions; update upgrade, HTTP, deprecation, article and catalog guidance. See the MB-820 report.

• September 28, 2026: deeply review MB-800’s 123 objectives; add seven worked examples, ten labs and 48 answered checks. Clarify security, inventory/ledger corrections and agent controls; curate useful articles and refresh learning resources. See the MB-800 report.

• September 28, 2026: deeply review MB-500’s 89 objectives; add six worked examples, ten labs, 48 answered checks and a locally executed replay/rollback model. Correct Batch OData and entity-security assumptions; refresh UDE/migration, X++ contracts, integrations and catalogs. See the MB-500 report.

• September 28, 2026: deeply review MB-310’s 96 published objectives, preserving the older condensed snapshot. Add eight worked examples, ten labs, 48 answered checks, current Finance behavior, three useful article tasks and refreshed catalogs. See the MB-310 report.

• September 28, 2026: deeply review MB-230’s 50 objectives; add six worked examples, ten labs, 48 answered checks and two useful blog tasks. Refresh SLA/capacity, knowledge access, agent previews, routing evidence, macro/survey behavior and overlapping learning paths. See the MB-230 report.

• September 28, 2026: deeply review DP-900’s 30 objectives; add six worked exercises, eight labs, 30 answered checks and two useful blog tasks. Fix SQL identity/grain examples, clarify Cosmos product boundaries and refresh Fabric, Power BI and learning catalogs. See the DP-900 report.

• September 28, 2026: deeply review AZ-900’s 57 objectives; add six worked examples, seven labs, 30 answered checks and three useful blog exercises. Refresh responsibility, billing, storage, identity and governance distinctions and learning catalogs. See the AZ-900 report.

• September 28, 2026: deeply review AZ-700’s 136 objectives; add six worked examples, ten labs, 48 explained checks and three Microsoft blog exercises. Update NAT/DNS/routing, VPN/ExpressRoute, private access, policy and lifecycle guidance, and learning catalogs. See the AZ-700 report.

• September 28, 2026: deeply review AZ-400's 86 objectives; add seven worked examples, ten labs, 48 explained checks and three Microsoft blog exercises. Update pipeline trust, artifact permissions, runtime/policy/retention changes, Azure DevOps identity and lifecycle, CodeQL, telemetry and learning catalogs. See the AZ-400 report.

• September 28, 2026: deeply review AZ-305's 49 objectives; add seven worked examples, ten labs, 48 explained checks and three bounded blog exercises. Refresh application recovery, APIM/Service Bus capabilities, current lifecycle changes and learning catalogs. See the AZ-305 report.

• September 28, 2026: deeply review AZ-140's 78 objectives; add seven worked examples, ten labs, 48 explained checks and three Microsoft blog exercises. Update imminent classic/client/Teams deadlines, managed host lifecycle, Hybrid, profiles, autoscale, servicing and learning catalogs. See the AZ-140 report.

• September 28, 2026: deeply review AZ-120's 78 objectives; add seven worked examples, ten labs, 48 explained checks and three Microsoft blog exercises. Refresh SAP storage, MANA, backup, operations, automation and learning catalogs; track the unresolved HSR backup procedure scope and October Sentinel connector image deadline. See the AZ-120 report.

• September 28, 2026: deeply review AZ-104's 82 objectives; add seven worked examples, ten labs, 48 explained checks and two Microsoft blog exercises. Update permissions, storage, deployment, encryption, app/certificate/backup, network and recovery guidance; refresh learning metadata and preserve review history. See the AZ-104 report.

• September 28, 2026: deeply review AI-901's 29 objectives; add six worked examples, eight labs, 36 explained checks, two Microsoft blog exercises and a Pluralsight implementation course. Update Foundry clients, speech/voice, vision lifecycle, Content Understanding and catalog evidence; preserve historical reviews and schedule follow-ups. See the AI-901 report.

• September 28, 2026: deeply review AI-300's 58 objectives; add six worked examples, ten labs, 48 explained checks and two Microsoft blog exercises. Update pipeline reuse, feature lineage, safe rollout, network/capacity, evaluation/telemetry, model lifecycle and learning catalogs; preserve historical evidence and schedule follow-ups. See the AI-300 report.

• September 28, 2026: deeply review AI-200's 27 objectives; add six worked examples, ten labs, 48 explained answers and a Redis engineering blog worksheet. Correct ACR/AKS identity, change-feed/retry, vector/cache, runtime deadlines, secret refresh, telemetry sampling and learning catalogs; retain the Redis conversion documentation gap with dated follow-ups. See the AI-200 report.

• September 28, 2026: deeply review AB-731's 39 objectives; add six business calculations, two artifact labs, 48 explained answers and a scoped Microsoft transformation blog exercise. Update value/adoption measurement, commitments, data/provider boundaries, governance and Image Analysis lifecycle. Preserve historical snapshots, synchronize catalogs and schedule two follow-ups. See the AB-731 report.

• September 28, 2026: deeply review AB-620's 44 objectives; add six worked examples, two labs, 48 answers and two blog exercises. Update course/language metadata, harness and Activity-protocol support, approval routing/ALM, retrieval, cards, evaluation, telemetry and throughput. Preserve prior review evidence, synchronize the learning catalog and schedule two follow-ups. See the AB-620 report.

• September 28, 2026: deeply review AB-410’s 48 objectives; add six examples, two labs, 48 answers and two focused Power Platform blog exercises. Explain prompt freshness, summary scope, generated-page ALM, query completeness, errors, approvals, credit transition and stale aggregates. Preserve historical snapshots and the SAP support gap; refresh learning catalogs and schedule three checks. See the AB-410 report.

• September 28, 2026: deeply review AB-250’s 119 objectives; add six examples, two labs, 48 answers and two WEM blog exercises. Expand telephony migration, agent limits, capacity, masking, outreach, forecasting, quality and diagnostics. Preserve conflicting ACS number eligibility/dates and historical snapshot evidence; schedule three lifecycle checks. See the AB-250 report.

• September 28, 2026: deeply review AI-500’s 55 objectives; add six examples, two labs, 44 answers and two engineering blog exercises. Correct hosted migration, A2A versions, retrieval filtering, interrupt replay and tool-support boundaries; qualify learning catalogs and schedule course/beta follow-ups. See the AI-500 report.

• September 28, 2026: deeply review AB-650’s 58 objectives; add six examples, two labs, 48 answers and a qualified Agent 365 blog exercise. Update Backup, network, meeting, risk migration, Cowork billing, agent identity, templates and Purview guidance; qualify new practice-test leads and preserve the second-year Backup cadence gap. See the AB-650 report.

• September 28, 2026: synchronize the shared learning catalog with completed guides, including previously missing exam entries. Add a read-only drift check and an explicit update command that preserves other entries and adjusts local links; include it in the review process and publication checks.

• September 28, 2026: deeply review AB-210's 46 objectives; add six worked examples, two labs, 44 answers and two qualified blog exercises. Correct transition/resource evidence and agent lifecycle, handoff, pricing, assignment, capacity and channel guidance. Preserve two official documentation contradictions with scheduled checks. See the AB-210 report.

• September 28, 2026: recheck the AB-100 prerequisite discrepancy; both conflicting lists remain published. Keep the blocker open and schedule October 5 follow-up without resetting the full-review receipt. See the bounded recheck.

• September 28, 2026: prevent recently reviewed blockers and previously reviewed event dates from repeatedly displacing initial deep reviews. Keep unresolved evidence gaps visible, publish the next eligible review date, and reopen scheduling for changed text, later due events or an expired interval. See the review process.

• September 28, 2026: deeply review SC-300's 98 objectives; preserve the unresolved official weighting discrepancy. Add five examples, two labs, 48 answers and two bounded blog exercises covering effective roles, baseline scopes, risk migration, downstream access removal, PIM timing and retained evidence. See the SC-300 report.

• September 28, 2026: deeply review MS-102's 54 objectives; add five worked examples, two labs, 44 answers and two blog exercises. Expand Backup, identity migration, policy precedence and DLP evidence; qualify catalogs and record the unresolved second-year Backup cadence. See the MS-102 report.

• September 27, 2026: deep-review MD-102's 83 October objectives while retaining July's baseline. Correct legacy app-grant guidance; add five examples, two labs, eight answers and two qualified blog exercises. Develop deployment/rollback, backup/restore, query freshness, hotpatch, remediation and agent lifecycle boundaries; qualify current catalog evidence and upcoming release entries. See the MD-102 report.

• September 27, 2026: deep-review DP-300's 94 October objectives while retaining April's baseline. Add five worked examples, two labs, eight answers and two qualified Microsoft blog readings. Fill Intelligent Insights coverage; clarify watcher targets, Query Store aggregation, enclave support, compaction, job identity and backup immutability. Refresh course metadata and schedule preview follow-ups. See the DP-300 report.

• September 27, 2026: deep-review SC-900's 58 October objectives while retaining July's baseline. Add four worked examples, two labs, eighteen answered checks and two qualified passkey blog exercises. Explain workload identities, access-review outcomes, network segmentation, threat intelligence, label publication and retention; qualify current lifecycle and commercial-resource claims. Add five follow-ups. See the SC-900 report.

• September 27, 2026: deep-review SC-200's 54 October objectives while retaining July's baseline. Add five worked examples, two labs, eight answers and two qualified Microsoft blog exercises. Update September Sentinel onboarding/history/Fabric paths, collection budgets, event identity, automation, NRT latency and resource metadata. Preserve two Microsoft documentation conflicts and schedule three follow-ups. See the SC-200 report.

• September 27, 2026: deep-review SC-100's 81 October objectives while retaining July's baseline. Correct the prerequisite list and agent policy targeting; add five worked examples, two labs, eight answers and two qualified Microsoft blog exercises. Clarify recovery, Sentinel migration, Key Vault APIs, CSPM/CIEM, network acquisition, Copilot DLP and malware-result boundaries; schedule five lifecycle checks. See the SC-100 report.

• September 27, 2026: deep-review MB-330 and map 101 October objectives while retaining June's baseline. Add four examples, two labs, 44 answers and two qualified Microsoft blog exercises. Clarify quality holds, reservation units, consignment, landed cost, mobile authentication/support and Planning Optimization; qualify stale commercial-resource claims. Update roadmap research guidance and add mobile/CTP/roadmap follow-ups. See the MB-330 report.

• September 27, 2026: deep-review AB-730 and map 41 October objectives while retaining July's baseline. Expand model, connector, notebook, meeting and Cowork preparation; add four worked examples, two labs, 44 answers, two Microsoft blog exercises and three rollout follow-ups. Correct current Learn/Pluralsight listings and qualify older paid-resource scope. See the AB-730 report.

• September 27, 2026: deep-review DP-800 and map 73 October objectives while retaining March's baseline. Correct CES scope, RLS update protection, DAB 2.0 defaults and external-model scope. Add five worked examples, two labs, 44 answers, two qualified Microsoft blog readings and three vector/CES follow-ups. See the DP-800 report.

• September 27, 2026: deep-review DP-750's 77 October objectives while retaining March's baseline. Correct ten-language availability and SQL paths/order; add four examples, two labs, 44 answers and two qualified Databricks blog readings. Clarify ABAC identities, replay/schema/quality, test isolation, direct-engine migration and maintenance windows; add three follow-up checkpoints. See the DP-750 report.

• September 27, 2026: deep-review PL-300's 78 April objectives and accept only the compared editorial snapshot, preserving its previous text/hashes. Add four examples, two labs, 44 answers, SQLBI performance reading, the February 2027 Q&A extension, and two maintenance checkpoints. Correct calculation, visual-export, RLS, refresh and Copilot boundaries, including a bounded DP-600 column clarification. See the PL-300 report.

• Preserve historical source reviews and exact objective snapshots when accepting a new baseline. Validate audits against the source review applicable to their date; retain past evidence without allowing new audits to use superseded baselines.

• September 27, 2026: deep-review DP-700 and map 54 October objectives while retaining July's baseline. Add reset/delete/CDF, time-policy, runtime, security and monitoring boundaries; four worked examples, two labs, 44 answers and one qualified Copy job blog exercise. Add a Runtime 2.0 rollout recheck. See the DP-700 report.

• September 27, 2026: deep-review DP-600 and map 41 October objectives while preserving July's baseline. Clarify Direct Lake variants, OneLake grants, Warehouse keys, export/freshness, RLS and incremental-refresh boundaries. Add four examples, two labs, 44 answers, two qualified blog readings and three planned ADBC follow-ups. See the DP-600 report.

• September 27, 2026: deep-review PL-400 and map 88 October replacement objectives while preserving March's baseline. Correct pipeline, connector, transaction, synchronization and Upsert boundaries; add managed-identity version 2 and explicit code-app/API/agent transition preparation. Add four examples, two labs, 44 answers and one qualified developer blog. See the PL-400 report.

• September 27, 2026: deep-review SC-401's whole guide and 65 published October objectives; retain July's baseline and the conflicting official revision dates. Correct Explorer/JIT boundaries and add Power BI, retention, audit and network DLP examples, a ninth lab, 42 answers and one qualified Purview blog. See the SC-401 report.

• September 27, 2026: deep-review AB-900's complete guide and 53 published October objectives while retaining the July baseline. Add current sharing/revocation, scheduled-prompt, restricted-site, PAYG and pending-version boundaries; four examples, a seventh lab, 14 answers and one qualified Mechanics reading. See the AB-900 report.

• September 27, 2026: deep-review DP-420's complete guide and all 56 published October objectives while preserving the current July baseline. Expand retrieval, memory, masking, Agent Kit, copy-job, Fleet and Fabric preparation; correct change-feed bootstrap, TTL, pagination and backlog drain. Add four examples, two labs, 36 answer checkpoints and two qualified blog readings. See the DP-420 report.

• September 27, 2026: deep-review AZ-802's 115 objectives; correct second-hop, dMSA, GPU-P, File Sync, SMB and monitoring lifecycle boundaries, and remove stale beta wording. Add four worked examples, a ninth lab, six answered checks and two qualified blog readings. Retain the unresolved SSH Direct support gap and training/practice availability discrepancies. See the AZ-802 report.

• September 27, 2026: deep-review all 115 AZ-801 objectives; add current Kerberos RC4 enforcement, OSConfig restoration limits, time-based backup immutability and retired direct telemetry guidance. Add four worked learning examples, a ninth lab, six answered checks and one qualified blog reading; preserve the canonical-blueprint weight discrepancy and dated lifecycle boundaries. See the AZ-801 report.

• September 27, 2026: deep-review AZ-800's 101 detailed objectives; correct Entra VM sign-in scope, dMSA prerequisites, PowerShell second-hop limitations and SMB defaults, and add five worked examples, a tenth lab and five answered checks. Add two qualified blog readings, preserve unresolved SSH Direct/Conditional Access warnings and refresh the shared SSH evidence. Publish a Microsoft review tracker with evidence-backed receipts, automated work packets and a documented per-certification validation/commit/push process. See the AZ-800 report.

• September 27, 2026: complete the GH-300 and AI-103 deep reviews across 105 detailed objectives. Correct Copilot policy/default and IDE-support boundaries, repair the required-check workflow example, clarify Foundry identity migration and evaluation, and add four curated blog readings, worked examples, three labs, and answered checks. Register 14 new sources, associate two existing sources, and add three dated Copilot maintenance follow-ups. See the deep-review report.

• September 27, 2026: complete the AB-100 deep research and guide repair pass. Clarify the October editorial revision and conflicting official prerequisites; add runtime-specific Copilot Studio guidance, trigger/MCP/computer-use identity boundaries, Foundry routing and publishing-model distinctions, preview agent-feed privacy and approval limits, a twelfth exercise, and answer checkpoints for all 17 original questions. Record sources, review limits, and remaining vendor conflicts in the deep-review report.

• September 27, 2026: sweep all 223 published guides, 42 catalog/announcement sources and 3,374 registered links (including six new citations); record exact per-guide outcomes, retries, 28 unchanged public PDF files and unresolved access limits. Update 20 guides with 12 October Microsoft revisions, AB-730 agent/Cowork preparation, DP-420 and PL-400 transition maps, active AZ-802 delivery, AB-100 Expert naming, announced Security+ SY0-801, FortiSASE Japanese delivery, and GitHub naming/release notes.

• Add one-command and Monday GitHub Actions maintenance reporting, 27 dated review tasks, candidate objective diffs, 90-day evidence retention, and a recurring review issue. Preserve accepted baselines and historical review dates, detect partial/stale runs, expose catalog title changes, and monitor CompTIA version selectors and Duration labels.

• Publish a partner-safe reference for the publicly named Claude Certified Architect, Foundations credential and its solution-architect audience; map public first-party material across requirements, deployment surfaces, Messages, prompting/context, retrieval/tools/MCP, agents, evaluation, security and operations without reconstructing Partner Academy objectives, with three scenarios, eight labs, 28 checks, a private-overlay boundary and weekly promotion triggers.

• Publish a dated OpenAI AI Foundations partner reference and public-first-party learning map covering responsible use, prompting, Responses API, structured outputs, retrieval, tools, agents, evaluation, safety, data controls and production practice; explicitly record the absent public blueprint/delivery/practice/renewal contract, keep it out of exam counts and CERTIFICATIONS.txt, add 24 readiness checks, eight evidence labs and automated maintenance triggers.
• Close the selected Microsoft phase at 50 published exam guides after a September 1 lifecycle reconciliation; document that MB-335/MB-700/PL-500/PL-600, AZ-204/MB-280, and AZ-500/PL-200 are recent retirements rather than missing current targets, while retaining the covered earnable AZ-800/AZ-801/MS-102 transitions.
• Publish a source-validated 4,059-word MB-820 guide mapped to all six June 10, 2025 domains, with platform/System/Base/extension and AppSource boundaries, reproducible AL environment/lifecycle, complete object and language patterns, least-privilege permissions, automated tests/telemetry/performance, resilient REST/API integration, three scenarios, eight labs and 36 checks; explicitly warn about the older baseline, record the active 100-minute seven-language exam, 50 hours 7 minutes of selected official paths, five-day course, MIT labs, free Practice Assessment and vetted commercial/live supplements.
• Publish a source-validated 3,774-word MB-800 guide mapped to all four June 30, 2026 domains, with company migration/security/automation, dimensions/workflows, accounting and posting-group controls, receivables/payables/assets, trade configuration, full document/payment/inventory lifecycles, three scenarios, eight labs and 36 checks; record the June blueprint changes, active 100-minute eight-language exam, 43 hours 40 minutes of official paths, five-day course, MIT labs, free Practice Assessment, current commercial supplements and partner access.
• Publish a source-validated 3,200-word MB-500 guide across all seven January 30, 2026 domains, with extension-first X++/AOT development, UDE/PPAC/LCS/Implementation portal boundaries, ALM, reporting/integration/security/performance contracts, three scenarios, eight labs and 36 checks.
• Publish a source-validated 3,500-word MB-330 guide mapped to every June 20, 2025 objective, with product/inventory, quality/assets, procurement/sales/landed cost, warehouse/mobile/transport and planning execution, three integrated scenarios, eight labs, 36 checks and freshness boundaries for the year-old blueprint.
• Publish a source-validated 3,900-word MB-310 guide mapped to every August 14, 2026 objective, with source-document-to-ledger accounting flows, three integrated scenarios, eight labs, 36 original checks, 14 exact learning choices, and explicit warnings for removed cost-management content and stale commercial practice material.

Notable public changes are recorded here. The project uses semantic version tags for release checkpoints; guide freshness and exam status remain independently dated in the exam catalog and each guide.

2026-09-28 — MS-721 deep review

• Map all 173 unchanged objectives; expand Teams Phone routing, Shared Calling, SMS, emergency-location and Rooms operational boundaries.
• Add six worked examples, two labs and twelve answers, with current Android management/AOSP, interop and XML guidance.
• Qualify blogs and commercial metadata, track dated follow-ups, and record offline-only validation.

2026-09-28 — MS-700 deep review

• Map 97 upcoming objectives while preserving the current baseline; refresh Teams administration, events, voice, app migration and diagnostics.
• Add five worked examples, two labs and eight answered checks; qualify older paid resources and retain the private-channel Workflows source conflict.
• Add actionable first-party blog readings and dated follow-ups; record source access and local validation evidence.

Unreleased

• Add a recurring, review-gated official-source freshness workflow and its first GitHub/Microsoft/Azure scan; queue 23 first-party candidates, correct three GH-300 lifecycle/capability statements, synchronize AB-650's current display name, and recognize a redirected Copilot Studio source's canonical harness overview.
• Repair and fresh-context revalidate the GitHub/Microsoft/Azure audit wave: correct GitHub lab safety, Actions YAML, security-campaign, and Copilot-plan coverage; deepen Agent 365, business-agent architecture, Agent Builder, Intune alerting, Azure hybrid networking, Sora remix, and Content Understanding workflows; register and health-check 22 first-party sources; resolve 22 of 24 findings; and retain honest SSH Direct blockers for AZ-800 and AZ-802 rather than inventing undocumented implementation guidance.
• Complete the selected NVIDIA expansion with a 4,072-word source-validated NCP-AIO performance guide across Installation and Deployment (31%), Administration (23%), Workload Management (23%), and Troubleshooting and Optimization (23%). Cover BCM/Mission Control/DOCA, Slurm/Kubernetes/Run:ai/MIG, NGC training/inference and cross-layer Docker/fabric/BCM/Magnum IO/storage diagnosis through a safe operator loop; add three scenarios, eight timed labs, 48 checks and ten reachable learning/reference links; preserve the 30-question plus three-integrated-lab, 120-minute contract.
• Publish a 3,651-word source-validated NCA-AIIO guide across Essential AI Knowledge (38%), AI Infrastructure (40%) and AI Operations (22%). Connect workload/CPU/GPU/software reasoning to topology, fabric, storage, facility, deployment, scheduler/orchestrator, telemetry, diagnosis and sharing decisions; add three scenarios, eight safe labs, 40 checks and eight learning choices; preserve NVIDIA's USD 125 certification-page versus USD 135 learning-path-card discrepancy.
• Publish a 2,918-word source-validated NCA-GENL guide across Core ML and AI Knowledge (30%), Software Development (24%), Experimentation (22%), Data Analysis and Visualization (14%), and Trustworthy AI (10%). Connect governed use-case/data/model/RAG/prompt/application choices to controlled experiments, deployment/monitoring and risk evidence; add three scenarios, eight labs, 40 checks and eight learning choices; preserve NVIDIA's conflicting 50 and 50–60 question statements rather than inventing a resolution.
• Inventory the selected NVIDIA expansion from the live certification catalog and weighted blueprints: NCA-GENL Generative AI LLMs, NCA-AIIO AI Infrastructure and Operations, and hands-on NCP-AIO AI Operations. Add exact query seeds, vendor monitoring, objective/status extraction and a unit test; preserve NVIDIA's associate/professional, prerequisite, two-year validity and integrated-lab distinctions before guide publication.
• Complete the four-guide ISC2 expansion with a 3,758-word source-validated CISSP guide across the active April 15, 2024 eight-domain 16/10/13/13/13/12/13/10 blueprint. Connect mission, ethics/law/governance, assets/risk, architecture and people choices, accountable decisions, assessment/operational evidence and recovery/improvement; integrate AI assets across domains; add three cross-domain scenarios, eight labs, 40 checks and 11 learning choices; distinguish the exam from experience/waiver/Associate, endorsement, ethics, 120-CPE cycle and AMF.
• Publish a 3,636-word source-validated ISC2 CCSP guide against the August 1, 2026 outline across cloud architecture/design (17%), data (20%), platform/infrastructure (17%), applications (16%), operations (17%) and legal/risk/compliance (13%). Connect obligation/data, service model/roles, responsibility/trust, control/contract evidence and recovery/exit; integrate new AI/ML and LLM/cloud-native scope; add three scenarios, eight labs, 40 checks and ten learning choices; distinguish exam readiness from experience/waiver/CISSP/Associate, endorsement, ethics, 90-CPE cycle and AMF.
• Publish a 3,024-word source-validated ISC2 SSCP guide against the October 1, 2025 outline: Security Concepts and Practices (16%), Access Controls (15%), Risk Identification/Monitoring/Analysis (15%), Incident Response/Recovery (14%), Cryptography (9%), Network/Communication Security (16%), and Systems/Application Security (15%). Add the asset-to-evidence operator loop, three scenarios, eight authorized labs, 40 checks and nine learning choices; explicitly separate exam readiness from experience/degree waiver, Associate status, endorsement, ethics, the 60-CPE three-year cycle and current AMF.
• Publish a 3,608-word source-validated ISC2 CC guide against the new September 1, 2026 outline: Security Principles (24%), Security Governance (17.3%), IAM Concepts (20%), Networking and Cloud Security Concepts (21.3%), and Security Operations/Incident Response (17.3%). Add connected risk/control/evidence reasoning, embedded AI-security context, three scenarios, eight authorized labs, 40 checks and nine learning choices; explicitly separate the no-experience exam from membership, ethics, 45-CPE/three-year and current AMF requirements.
• Inventory the selected ISC2 expansion from current first-party outlines: the September 1, 2026 CC revision, October 1, 2025 SSCP baseline, August 1, 2026 CCSP revision, and April 15, 2024 CISSP baseline. Add an ISC2 objective adapter and extraction test, exact query seeds, and an explicit exam-pass versus certification experience/endorsement boundary before publishing the four guides.
• Publish a 4,042-word source-validated GES-C02 SnowPro Specialty: Gen AI guide across all four live public abilities: Gen AI principles and responsible/evaluated architecture; current Cortex AI functions, RAG, semantic views, Agents and MCP; reproducible open-model tuning and Model Registry/SPCS lifecycle; and governed, idempotent document parse/extract/chunk/index pipelines. Add three scenarios, eight labs, 40 checks and 11 reachable learning/reference links while preserving the form-delivered detailed-guide boundary without inventing inaccessible weights or subobjectives.
• Publish a 3,292-word source-validated DEA-C02 SnowPro Advanced: Data Engineer guide across all five live public abilities: lake/API/on-premises sourcing, transformation/replication/sharing, near-real-time streams, scalable compute and performance evidence. Add production semantics for contract/reconciliation/idempotency/replay/backpressure/SLOs, three integrated scenarios, eight labs, 40 checks and ten timed learning choices; retain the two-year production-experience recommendation and gated detailed-guide boundary without inventing weights.
• Publish a 3,694-word source-validated COF-C03 SnowPro Core guide across all seven live public abilities: architecture, accounts/warehouses, load/unload/transform, data forms, performance, collaboration/protection and connectivity. Add three integrated scenarios, eight evidence labs, 40 checks and ten timed/scoped learning choices; record the six-month recommendation, gated detailed-guide boundary, active official practice, current policies and COF-C02 freshness risk without inventing inaccessible subobjectives or weights.
• Publish a 3,006-word source-validated retired SOL-C01 SnowPro Associate: Platform reference across its six preserved public abilities: account/UI/notebook context, objects/stages/compute, structured/semi-structured/unstructured loads, roles/access, protection/sharing, and Cortex AI SQL. Add three scenarios, eight labs, 36 checks and seven reachable sources; direct new learners to the free non-expiring Platform Skills Badge, preserve the badge-versus-SnowPro distinction, and freeze retired objective baselines when their former landing pages disappear.
• Inventory the four-page Snowflake expansion against live first-party lifecycle pages: retired SOL-C01 Associate: Platform and its May 5, 2026 free non-expiring Snowflake University Platform Skills Badge replacement; active COF-C03 Core; active DEA-C02 Advanced: Data Engineer; and active GES-C02 Specialty: Gen AI, which replaced GES-C01 on May 19 before the old version retired July 20. Add exact machine-readable seeds, vendor monitoring and extraction tests so retired SOL-C01/COF-C02/GES-C01 cannot be presented as current.
• Complete the four-guide Cisco expansion with a 4,043-word source-validated 200-901 CCNA Automation guide mapped to all six 15/20/15/15/20/15 v1.1 domains: data formats/Python/Git/tests/design, documented API construction and diagnosis, Cisco platforms/SDKs/YANG/NETCONF/RESTCONF, deployment/Docker/CI-CD/Bash/security, reviewed IaC/Ansible/Terraform/NSO automation, and application-aware networking. Add three scenarios, eight authorized labs, 46 checks and 11 learning/reference links; explicitly reconcile the DevNet rename, v1.0/v1.1 page discrepancy and inconsistent first-party language listings.
• Publish a 3,775-word source-validated 200-301 CCNA guide against the live six-domain v1.1 blueprint through February 2, 2027, with packet/control-plane reasoning, minimum configuration, verification, faults, correction and rollback across fundamentals, access, routing, services, security and automation; separately map every scheduled v2.0 domain and material delta for February 3 without blending future and current scope; add three scenarios, eight authorized labs, 36 checks and ten timed or scoped learning choices.
• Publish a 3,541-word defensive, source-validated 100-160 CCST Cybersecurity guide across all five public work areas and 23 current first-party training objectives. Connect asset/threat/vulnerability/risk/control reasoning to identity and cryptography, TCP/IP/network/wireless defense, endpoint baseline/log/update/malware procedures, authorized vulnerability assessment/contextual prioritization, continuity/recovery, evidence/chain-of-custody and incident escalation/response; add three scenarios, eight safe labs, 36 checks and nine learning links, including Cisco's free approximately 120-hour path, while omitting unavailable stable domain weights.
• Publish a 3,973-word source-validated 100-150 CCST Networking guide across all six public topic groups and every current first-party training objective. Build packet-delivery reasoning through models/performance/protocols, IPv4/IPv6 and subnetting, endpoints/copper/fiber/wireless, switching/routing/services/read-only IOS evidence, controlled diagnosis/Wireshark/ticketing, and foundational firewall/WPA/security practice; add three scenarios, eight authorized labs and 36 checks, explicitly omit unavailable stable domain weights, and catalog ten official/learning links with a free approximately 70-hour Cisco path and transparent book/course estimates.
• Inventory the four-guide Cisco expansion set from current first-party pages: 100-150 CCST Networking, 100-160 CCST Cybersecurity, 200-301 CCNA, and 200-901 CCNA Automation. Add the Cisco vendor and monitor adapter, exact machine-readable seeds/list rows, extraction tests, and the scheduled CCNA v1.1-to-v2.0 boundary (last v1.1 testing February 2, 2027; v2.0 live February 3) before guide authoring.
• Publish a source-validated Professional Agentic Architect beta guide one day before its announced September 3 registration opening. Map all five 13/17/33/22/15 domains and the two-part exam/lab contract across low-code and coding agents, ADK, memory/RAG/tools/MCP/A2A/multi-agent workflows, layered evaluation, runtime/trace/AgentOps and OAuth/PAB/gateway/registry/Model Armor/HITL governance; add three scenarios, eight labs, 36 checks and a 44-hour official path while explicitly recording beta volatility and the absent credible commercial exam market.
• Publish a source-validated Professional Machine Learning Engineer guide against the actual linked June 1, 2026 _new.pdf, not the different older resolving file. Cover low-code/model/API choice, governed features/notebooks/experiments, training/tuning/accelerators, batch/online serving and rollout, pipelines/CI-CD-CT, predictive/gen-AI quality, drift/skew, responsible/security controls; add three scenarios, eight labs, 36 checks and an explicit Vertex-to-Agent-Platform translation.
• Publish a defensive, source-validated Professional Cloud Security Engineer guide across the current 25/22/23/19/11 domains. Cover identity/privilege, NGFW/private/perimeter paths, sensitive-data/key/secret/AI protection, secure supply chain/posture, logs/detection/response and compliance evidence; add three scenarios, eight authorized labs, 36 checks, seven sources and a current-control gap checklist.
• Publish a source-validated Professional Data Engineer guide across the current 22/25/20/15/18 domains. Cover governed contracts, batch/stream correctness, current storage/warehouse/lake/platform choices, BI/ML/RAG preparation, sharing, orchestration/CI-CD, capacity/cost, observability, restart/replay and recovery; add three scenarios, eight labs, 36 checks, eight learning choices and an explicit old-course/current-scope gap list.
• Publish a source-validated Professional Cloud Architect guide across the current 25/17.5/17.5/15/12.5/12.5 domains and four V6.1 cases. Connect all six Well-Architected pillars to requirements, platform/data/network/AI choices, migration, security/compliance evidence, technical/business processes, implementation and operational excellence; add a reusable case method, three scenarios, eight labs, 36 checks and a current AI/agent/security gap checklist.
• Publish a source-validated Associate Cloud Engineer guide against the current four-domain 20/30/30/20 PDF and its branding-update notice. Cover hierarchy, policy, billing, quotas, networking, compute/GKE/Cloud Run/agents, data/storage, IaC, operations, observability, backup/recovery, IAM and federation; add three scenarios, eight labs, 40 original checks, eight cataloged learning/reference choices and an explicit old-outline/current-product gap checklist.
• Publish a 5,004-word source-validated Generative AI Leader guide across the current 30/35/20/15 domains, with foundation/data/model/layer decisions; Gemini, enterprise search, customer engagement, Agent Platform, RAG, APIs and agent tools; prompting/grounding/customization/evaluation; secure/responsible adoption and value; three scenarios, eight labs, 40 original checks and an explicit fast-moving 2026 product-name map.
• Publish a 5,733-word source-validated Cloud Digital Leader guide against the August 12, 2026 six-domain baseline. Cover outcome-led transformation, service/deployment boundaries, governed data and current data services, generative and agentic AI decisions, migration/compute/API choices, layered AI/cloud security, FinOps/observability/SRE, three scenarios, eight labs, 36 original checks and a precise pre-August gap checklist; register the first Google Cloud guide and learning/practice choices.
• Publish a source-validated 5,271-word MLA-C02 beta guide against the complete September 1 launch blueprint and official C01 comparison. Cover governed multimodal data, embeddings/vector stores/RAG preparation, traditional models and FM selection/customization, retrieval/generation/human/judge evaluation, SageMaker and Bedrock deployment, knowledge/index refresh, agent tools/protocols/state/versioning, MLOps/LLMOps testing, observability, tokens/vector/agent cost and layered AI security; add three scenarios, eight labs and 42 original checks. Record the ME1-C02 beta appointment code, September 29 delivery, 170-minute/85-question/USD-75 contract, early-2027 GA plan, and immature commercial-learning boundary.
• Publish a source-validated 5,153-word MLA-C01 guide across all four traditional-ML/MLOps domains, with data contracts, leakage-safe feature lifecycle, algorithm/training/tuning/evaluation decisions, inference and orchestration patterns, drift/operations/cost/security depth, three scenarios, eight labs and 40 original checks. Prominently record the September 28, 2026 English retirement, MLA-C02 replacement, ME1-C02 beta delivery code, and the expanded generative-AI/agentic boundary so new learners are not directed into a retiring version.
• Publish the 4,847-word source-validated AIF-C01 AWS Certified AI Practitioner guide against revision 1.0 dated March 26, 2026. Map all five domains to technique and service selection, model/data lifecycle, tokens/context, RAG/customization/prompting/evaluation, responsible AI, security and governance; make the new agentic AI, MCP, multi-agent, Amazon Quick, Kiro, Strands Agents and Bedrock AgentCore boundary explicit; add three integrated scenarios, eight labs, 35 original checks and current learning/practice choices with version gaps and time estimates.

• Publish the 5,780-word source-validated CLF-C02 AWS Certified Cloud Practitioner guide against the current four-domain official guide, with cloud-value and economics decisions; Well-Architected and Cloud Adoption Framework context; shared responsibility, identity, governance, monitoring and protection boundaries; global infrastructure and compute/storage/database/network/service-category selection; pricing, cost-management and support distinctions; three integrated scenarios, eight safe labs and 30 original checks. Catalog all guide sources and learning options with access and time estimates, establish the AWS website/provider pattern, and reduce the actionable first wave to 30 guides.

• Reconcile and register the 31-guide AWS, Red Hat, CompTIA, and Linux Foundation first wave from exact official sources. Expand AWS from the stale 12-guide assumption to 14 by adding AIB-C01 and tracking both retiring MLA-C01 and MLA-C02 beta; split CompTIA A+ V15 into its independently published 220-1201 and 220-1202 component guides; record five version-aware Red Hat performance anchors and five Linux Foundation/CNCF targets. Add four public-objective adapters with extraction tests and correct the first-wave roadmap target from 95 to 98 including the conditional OpenAI slot.

• Surface OpenAI AI Foundations and Anthropic Claude Certified Architect, Foundations on both the homepage and complete catalog with explicit provisional/public-blueprint and partner-gated states, while keeping them outside objective-mapped guide counts.
• Promote Partner AI references to a top-level site section with direct navigation to the provisional OpenAI AI Foundations and Anthropic Claude Certified Architect reference pages.
• Add a first-class Partner learning journeys area and a dated Frontier Transformation Engineer page that connects GH-300, AI-103 and AB-100 without double-counting a new certification. Record the public 11-course/~86-hour/three-milestone baseline; the active AI-102 substitution boundary; five Project Ready assessments; required Hypervelocity assessment; durations, 80% threshold, retake and synchronization notes; work-email/Learn/Partner Center setup; and the public-versus-gated content boundary. Register the program source, expose the pages in generated navigation, and test the publication allowlist.
• Complete the seven-certification Databricks catalog with a 4,758-word source-validated Databricks Certified Associate Developer for Apache Spark guide against the October 30, 2025 live-version PDF and current seven-domain page. Cover driver/executor/resources, job-stage-task/lazy/shuffle/cache/GC/fault-tolerance reasoning; schema-aware SQL/file/JDBC I/O and table/view/mode/layout behavior; DataFrame columns/nulls/dedup/aggregates/dates/joins/unions/I/O/UDF/shared variables; partition/skew/AQE/UI/log diagnosis; streaming state/watermark/checkpoint/recovery; Spark Connect versus deployment mode; and Pandas API/Pandas UDF boundaries. Add three scenarios, eight labs, 39 checks and 11 cataloged guide links, explicitly distinguish the active exam from the retired Spark 3.0 format, and correct a moved Spark 4.2 pandas-documentation route during source health validation.
• Publish the 4,946-word source-validated Databricks Certified Generative AI Engineer Associate guide against the March 18, 2026 live-version PDF and current six-domain page, with requirements/prompts/tasks/chains/tools/Agent Bricks; governed extraction/chunking/embeddings, AI Search/Vector Search, retrieval metrics and reranking; model/framework selection, layered guardrails, traced Agent Framework and Genie/multi-agent design; application packaging, registration, batch/interactive inference, persistent memory, CI/CD, prompt lifecycle, MCP and secure Apps; Unity Catalog/masking/licensing controls; evaluation, judges/scorers, monitoring, cost and SME feedback. Add three scenarios, eight labs, 37 checks and 12 cataloged guide links with explicit volatility boundaries for agent, MCP, Apps, AI Search, MLflow and AI Gateway interfaces.
• Publish the 3,091-word source-validated Databricks Certified Machine Learning Professional guide against the September 30, 2025 live-version PDF and current three-domain page, with Spark/single-node/Ray and vertical/horizontal/model/data/trial/group scaling, Optuna/nested MLflow/PyFunc, point-in-time/online/on-demand features, deploy-code environment architecture, complete unit/integration gates, governed automated retraining, sliced drift/performance/health monitoring and alerts, and blue-green/canary/custom-model serving; add three scenarios, eight labs, 32 checks and 17 cataloged links with current bundle/monitoring terminology boundaries.
• Publish the 3,471-word source-validated Databricks Certified Machine Learning Associate guide against the March 1, 2025 live-version PDF and current four-domain page, explicitly reconciling ML Workflows/Data Processing labels and live-versus-PDF delivery metadata; cover MLOps/runtime/AutoML, governed feature/MLflow/model lifecycle, leakage-safe processing, algorithm/pipeline/imbalance/tuning/CV/metric choices and batch/stream/realtime deployment, with three scenarios, eight labs, 32 checks, 16 cataloged links and current-name boundaries for Lakeflow, Unity Catalog registry, Hyperopt and serving.
• Publish the 4,491-word source-validated Databricks Certified Data Engineer Professional guide against the July 3, 2026 detailed guide and live ten-domain page, with modular Python/package/bundle design, dependency/UDF/testing boundaries, batch/stream/Lakeflow/AUTO CDC control, quality quarantine, D2D/open sharing and federation, layered system/query/Spark/event observability, measured cost/performance and CDF, privacy/purge/governance, repair/CI/CD and scalable Delta/dimensional modeling; add three production scenarios, eight labs, 37 checks and 30 cataloged guide sources with explicit commercial-discovery limits.
• Publish the 5,124-word source-validated Databricks Certified Data Analyst Associate guide against the October 30, 2025 detailed PDF and live nine-domain page, with platform/Catalog Explorer/Marketplace decisions; governed cleanup and intake; ANSI SQL, warehouse, join/set/aggregate/object/federation/history depth; evidence-led query analysis; AI/BI dashboards and Genie grounding/validation/maintenance; dimensional and Data Vault modeling; layered Unity Catalog security; three scenarios, eight labs and 36 checks. Extend the adapter to both Databricks coverage-heading variants and catalog 27 guide sources with duration and freshness boundaries.
• Publish the 4,932-word source-validated Databricks Certified Data Engineer Associate guide against the May 4, 2026 detailed PDF and live seven-domain weighted page, with platform/compute decisions; five ingestion patterns and schema/replay semantics; bronze/silver/gold transformation, joins, quality and tuning; Lakeflow Jobs control flow/triggers; Git folders and Declarative Automation Bundles; evidence-led run/Spark diagnosis; Unity Catalog lifecycle/privileges/fine-grained controls; three scenarios, eight labs and 36 checks. Add the Databricks objective adapter pilot, 28 cataloged learning/product sources with duration/freshness boundaries, the first Databricks website/provider entry, and a stable library identifier explicitly distinguished from a vendor exam code.
• Publish a source-validated 3,454-word MB-230 guide mapped to all three March 11, 2026 domains, with secure case lifecycle/automation/timeline, governed internal/translated/external knowledge, Teams collaboration, filtered Copilot and controlled Case Management Agent/tool actions, SLA/item/KPI/KPI-instance and Power Automate failure semantics, basic/unified routing/capacity/skills/diagnostics, scripts/slugs/macros, workspace templates/profiles/Inbox, Dataverse/UI/search/template/notification extension, Customer Voice correlation/governance, three scenarios, eight labs and 36 original checks; record the stale October 2025 credential-page discrepancy, active 100-minute seven-language exam, 29 hours 2 minutes of selected official paths, four-day course, free Practice Assessment, current Pluralsight and Udemy supplements, partner access and commercial-catalog gaps, and expand the monitored catalog to 1,719 sources.
• Publish a source-validated 3,561-word AB-250 guide mapped to all six domains on the official page last updated May 15, 2026, with standalone/embedded/third-party CCaaS choices, connectors/agents/ALM/users/capacity, chat/digital/custom/mobile and voice/IVR/telephony lifecycles, secure proactive engagement and WFM, grounded Copilot and voice-agent orchestration, unified routing/diagnostics, representative profiles/templates/productivity/knowledge, responsible supervisor/quality operations, Power BI and Application Insights, three scenarios, eight labs and 36 original checks; record the active 120-minute English exam, missing Practice Assessment, 11 hours 40 minutes of official paths, three-day course, LinkedIn Learning caveat, MB-240 non-direct partner transition, commercial-catalog gaps, and expand the monitored catalog to 1,704 sources.
• Publish a source-validated 3,732-word AB-210 beta guide mapped to all five domains on the official page last updated June 18, 2026, with lead-to-cash data/process/security readiness, mailboxes/Microsoft 365 collaboration, product/pricing/currency, AI-first data/reporting/plan decisions, accelerator/assignment and intelligence configuration, distinct Qualification/Opportunity/Close/Research agent contracts, research-only versus research-and-engage autonomy, forecasts/goals, mobile/calling/SMS and Power Platform extension boundaries, three scenarios, eight labs and 36 original checks; record the 120-minute English beta exam, delayed results, unavailable Practice Assessment, 12 hours 3 minutes of official paths, three-day course, MB-280 partner transition, current Udemy options, commercial-catalog gaps, and expand the monitored catalog to 1,694 sources.
• Publish a source-validated 4,688-word AB-731 guide mapped to all three July 22, 2026 domains, with AI/ML/automation and model/adaptation choices, prompt/RAG/data/ML-lifecycle/security decisions, credible token/ROI/opportunity economics, Copilot/Graph/Copilot Studio and build/buy/extend mapping, current Microsoft Foundry/Foundry Tools/Vision/Search boundaries, responsible-AI controls, AI-council/adoption-team/champion/workload ownership, licensing/consumption and staged adoption evidence, three scenarios, eight labs and 36 original checks; record the active 45-minute exam, 4 hours 44 minutes of official paths, one-day course, free Practice Assessment, official prep session, current Pluralsight/Udemy choices, commercial-catalog gaps, and expand the monitored catalog to 1,683 sources.
• Publish a source-validated 4,100-word AB-730 guide mapped to all three July 22, 2026 domains, with context/Graph/app choices, chat/agent/Researcher/Analyst/Page/Notebook boundaries, responsible-AI and data-protection decisions, goal-context-source-expectation prompts, save/schedule/share and conversation lifecycle, Agent Store/template/knowledge/settings/sharing, document/summary/cross-app/meeting/collaboration/memory workflows, three scenarios, eight labs and 36 original checks; record the active 45-minute exam, 4 hours 31 minutes of official learning, one-day course, free Practice Assessment, official prep video, current Pluralsight/O'Reilly/MeasureUp/Udemy options and freshness or page-quality caveats, and expand the monitored catalog to 1,672 sources.
• Publish a source-validated 5,300-word AB-410 guide mapped to all three domains on the official page last updated May 15, 2026, with requirement/component/agent decisions, environment governance and ALM, Dataverse models/security/prompt columns/row summaries, model-driven and accessible responsive canvas apps, reusable formulas/components/state/errors/Monitor, embedded-agent boundaries, reliable triggers/connectors/approvals/control/idempotency, governed prompts/knowledge/models/human review, business-rule/process/column decisions, three scenarios, eight labs and 36 original checks; record 15 hours 26 minutes of official paths, the three-day course, unavailable Microsoft Practice Assessment, course-versus-credential PL-200 transition boundary, verified and missing commercial options, and expand the monitored catalog to 1,662 sources.
• Publish a source-validated PL-400 guide mapped to all six March 19, 2026 domains, with extension/table/identity/security decisions, preview Power Fx functions, solution lifecycle and CI/CD, advanced Power Fx and diagnostics, Client API/commands/custom pages, PCF lifecycle, plug-ins/custom APIs/business events, OpenAPI connectors, Dataverse APIs/limits, Functions/managed identity, reliable cloud flows, event destinations and synchronization, three scenarios, eight labs and 36 original checks; record about 25 hours of official paths, the five-day course, MIT official labs, free Practice Assessment, current and dated commercial resources, credential-page staleness, and expand the monitored catalog to 1,650 sources.
• Publish a source-validated PL-300 guide mapped to all four April 20, 2026 domains, with source/storage/privacy/gateway decisions, Power Query profiling/folding/quality and dimensional transformation, star-schema relationships and DAX contexts, calculation groups and visual calculations, performance evidence, accessible report design and defensible analysis, Copilot validation, workspaces/apps/dashboards/distribution, refresh/RLS/labels, three scenarios, eight labs and 36 original checks; record 19 hours 52 minutes of current Learn paths, the three-day course, MIT official labs, free Practice Assessment, current commercial and public-video choices, freshness caveats, and expand the monitored catalog to 1,632 sources.
• Publish a source-validated SC-401 guide mapped to all three July 28, 2026 domains, with classification/SIT/fingerprint/EDM/classifier/OCR design, item and container labels, scanner and encrypted mail, cloud and Endpoint DLP, Adaptive Protection, retention and recovery, privacy-aware Insider Risk Management, Audit/Activity Explorer/XDR/eDiscovery investigations, current-versus-classic DSPM and AI controls, three scenarios, eight labs and 36 original checks; record 20 hours 45 minutes of current Learn paths, the four-day course, free Practice Assessment, current MeasureUp and dated Udemy coverage, and expand the monitored catalog to 1,608 sources.
• Publish a source-validated DP-700 guide mapped to all three July 21, 2026 domains, with Spark/domain/OneLake/Airflow settings, Git/database-project/deployment lifecycle, layered security/governance/audit, parameterized orchestration, full/incremental/dimensional/streaming loads, lakehouse/warehouse/Eventhouse/shortcut/mirroring decisions, PySpark/SQL/KQL transformation, named error isolation, evidence-led optimization, three scenarios, eight labs and 36 original checks; record 27 hours 49 minutes of current Learn paths, four-day course, free Practice Assessment and current commercial learning/practice options.
• Add explicit beginner, intermediate, and expert metadata to every published guide; group and naturally sort provider, all-guide, collection, and homepage listings by level and exam code; sort collection-card code summaries by the same rule; and show the editorial level on every guide card.
• Reorder production to complete Microsoft, then produce public-source-safe partner-oriented OpenAI and Anthropic coverage, then complete Databricks before a portfolio checkpoint; record Google Cloud, Cisco, Snowflake, ISC2, NVIDIA, Salesforce, MongoDB, ServiceNow, Palo Alto Networks, Fortinet, Splunk, ISACA, and Oracle as later expansion candidates.
• Add a public, checkable guide backlog for the 15 remaining Microsoft first-wave exams, grouped in production order with official study-guide links, lifecycle-review guidance, and explicit Office Specialist and Educator exclusions.
• Publish a source-validated DP-600 guide mapped to all three July 21, 2026 domains, with layered security/governance, Git/PBIP/deployment/XMLA lifecycle, lakehouse/warehouse/eventhouse/OneLake choices, dimensional transformation and reconciliation, SQL/KQL/DAX, semantic-model relationships/calculations/security/storage/performance, AI-ready data, three scenarios, eight labs and 36 original checks; record 23 hours 20 minutes of current Learn paths, four-day course, free Practice Assessment and dated commercial-resource gaps.
• Publish a source-validated MS-721 guide mapped to all four April 28, 2026 domains, with meeting/event and PSTN architecture, Audio Conferencing, Calling Plans/Operator Connect/Teams Phone Mobile/Direct Routing/Shared Calling, routing/SBC/SBA/LBR/LMO, auto attendants/queues/emergency calling, Rooms and device enrollment/security/lifecycle, flexible spaces, three scenarios, eight labs and 36 original checks; record 14 hours 43 minutes of current Learn paths, five-day course, free Practice Assessment and dated commercial-resource gaps.
• Publish a source-validated MS-700 guide mapped to all four July 29, 2026 domains, with media/network readiness, Teams data/security/compliance/governance, policy assignment, external/guest/shared-channel/MTO collaboration, Teams Rooms/VDI, teams/channels/apps, meetings/events/Copilot, Teams Phone call flows, monitoring/troubleshooting, three scenarios, eight labs and 36 original checks; record 19 hours 1 minute of Learn paths, four-day course, free Practice Assessment and freshness gaps in commercial resources.
• Publish a source-validated MD-102 guide mapped to all five July 24, 2026 domains, with device identity/enrollment/delegation/compliance, Autopilot and device preparation, Windows 365 and Intune Suite, remote action, endpoint security and updates, application/MAM lifecycle, Graph/PowerShell automation, Security Copilot agents, analytics/remediations/reporting, three scenarios, eight labs and 36 original checks; record 29 hours 46 minutes of current Learn paths, five-day course, free Practice Assessment and current-versus-older provider coverage.
• Publish a source-validated AB-650 beta guide mapped to all three domains on the official page last updated July 27, 2026, with tenant and workload operations, identity and Conditional Access, Defender for Office 365, Purview data governance, Copilot readiness and controls, agent identity/lifecycle/tool governance, Agent 365 security/cost/monitoring, three integrated scenarios, eight labs, and 36 original checks; record the 23 hours 8 minutes of official Learn paths, unavailable Practice Assessment and independently unverified third-party exam-specific offerings.
• Publish a source-validated 6,000-word MS-102 guide mapped to every April 28, 2026 objective, with tenant operations, user/group/license lifecycle, cross-workload delegation, Connect Sync and Cloud Sync, authentication and Conditional Access, Defender XDR/Office 365/Endpoint/Cloud Apps, Purview classification/retention/DLP, three integrated scenarios, eight labs, 36 original checks, and 56 exact references; prominently record the November 30, 2026 exam and certification retirement, absence of a named direct replacement, free Practice Assessment, 28 hours 45 minutes of official Learn paths, and expand the monitored catalog to 1,473 sources.
• Publish a source-validated 9,000-word SC-500 guide mapped to every objective on the official page last updated May 13, 2026, with PIM/Conditional Access/app and workload identity, Key Vault, Policy/compliance/RBAC/Backup/IaC, Storage/SQL/networking, AI and agent security, hybrid VM and application-platform protection, Defender posture, Sentinel collection and Security Copilot, three integrated scenarios, eight labs, 36 original checks, and 51 exact references; record the active credential, August 31 AZ-500 retirement, unavailable Practice Assessment, 30 hours 8 minutes of official paths, current versus legacy learning resources, fast-moving AI/Sentinel/product transitions, and expand the monitored catalog to 1,435 sources.
• Publish a source-validated 7,900-word SC-300 guide mapped to every April 27, 2026 objective, with tenant/workforce/external/hybrid identity, strong authentication and recovery, Conditional Access/risk/Global Secure Access, workload identities and enterprise applications, Defender for Cloud Apps, entitlement/access reviews/PIM/emergency access, KQL/monitoring, three integrated scenarios, eight labs, 36 original checks, and 45 exact references; disclose Microsoft's conflicting authentication-domain weight, current product transitions and older-resource gaps, record 15 hours 11 minutes of official Learn paths, and expand the monitored catalog to 1,395 sources.
• Publish a source-validated 9,200-word SC-200 guide mapped to every July 28, 2026 objective, with Defender XDR and Endpoint automation/custom data collection, AIR/attack disruption, Sentinel automation/roles/data tiers/workbooks/optimization, multisource ingestion, detection engineering, cross-domain incident response, KQL and Advanced Hunting, Sentinel Graph, data-lake jobs/summaries/notebooks/MCP, Security Copilot and agentic investigation, three integrated scenarios, eight labs, 36 original checks, and 48 cataloged references; record 43 hours 12 minutes of official paths, the free Practice Assessment, the March 31, 2027 Azure-portal transition, older-resource gaps, and expand the monitored catalog to 1,352 sources.
• Publish a source-validated 7,400-word SC-100 guide mapped to every July 28, 2026 objective, with ransomware/BCDR, MCRA/MCSB/Zero Trust/CAF/WAF, secure AI and DevSecOps, XDR/SIEM/SOAR, human/workload/agent identity, privileged access, compliance, multicloud posture/exposure, endpoints/OT/IoT/SSE, Microsoft 365 Copilot, application/API/WAF and data security, three integrated scenarios, eight labs, 36 original checks, and 36 cataloged references; record 20 hours 58 minutes of official paths, current prerequisites, the free Practice Assessment, older-resource gaps, and expand the monitored catalog to 1,306 sources.
• Publish a source-validated 7,000-word AB-620 guide mapped to every April 21, 2026 objective, with a classic-versus-new Copilot Studio boundary, architecture/identity/governance, flows/topics/responses/state, knowledge and tool integration, child/connected/Foundry/Fabric/A2A agents, evaluation/telemetry, solutions/pipelines, three scenarios, eight labs, 36 original checks, and 31 cataloged references; record the active credential, live self-paced paths, September 18 instructor-led course date, and expand the monitored catalog to 1,271 sources.
• Publish a source-validated 8,200-word AI-500 beta guide mapped to every July 16, 2026 objective, with a current-versus-classic Foundry boundary, architecture contracts, Agent Framework/HITL, MCP/A2A, per-agent identity, memory/RAG, tools, evaluation/operations, guardrails/red teaming, release engineering, three integrated scenarios, eight labs, 36 original checks, and 32 cataloged references; distinguish the live beta exam from the September 30 course date and expand the monitored catalog to 1,245 sources.
• Apply the September 1 public-site review: normalize every guide to one H1 and a major-section H2 hierarchy, limit guide TOCs to major sections, replace the expanding all-guide rail with provider-level navigation, add exam-essentials/deep-understanding/hands-on entry points, clarify that the sources-and-objectives gate is AI-assisted and human review remains pending, deep-link every guide to its evidence record, add a real maintainer/About page, repair the homepage skip link and heading whitespace, state the exact monitored-guide count, and document external review and search-indexing follow-ups.
• Refine CERTIFICATIONS.txt into a minimal Python-friendly query-seed table containing only vendor ID, exam code, and title, with a standard-library parsing example and a clear boundary between input identities and downstream enrichment results.
• Audit AI-103 against the current Microsoft Foundry documentation generation, add a current-versus-classic terminology and SDK crosswalk, replace legacy aliases and one genuinely classic tracing reference with canonical current sources, explicitly label classic third-party training, and expand the monitored catalog to 520 entries.
• Add a generated, tab-separated CERTIFICATIONS.txt inventory for downstream mirrors, with a regeneration command, mirror guidance, tests, and repository validation that prevents drift from the canonical exam catalog.
• Expand AB-900 into object, entitlement, identity, authorization, protection, admin-surface, and evidence decisions, with collaboration/access, sign-in, governed-document, grounded-response, oversharing, billing, measurement, agent-approval, and incident flows; promote it to SOURCE-VALIDATED, register 18 exact Microsoft sources, complete source validation for every cataloged guide, and expand the monitored catalog to 519 entries.
• Expand PL-900 from product definitions into requirement-to-solution, Dataverse/data, identity/connection, release, app request, flow execution, agent-turn, and estate-governance decisions; cover the July 2026 plans, code apps, vibe, agent flows, evaluations, and Agent 365 additions; promote it to SOURCE-VALIDATED, register 27 exact Microsoft sources, and expand the monitored catalog to 501 entries.
• Expand SC-900 into asset, identity, threat, control, evidence, and response decisions, with worked identity-request, inbound-traffic, security-signal, and data-lifecycle flows plus an integrated objective drill; promote it to SOURCE-VALIDATED, register 33 exact Microsoft sources, and expand the monitored catalog to 474 entries.
• Deepen AB-100 with direct architecture evidence and a cross-platform responsibility chain spanning experience, orchestration, models, grounding, actions, systems of record, operations, and identity; promote it to SOURCE-VALIDATED, register 11 exact Microsoft sources, and expand the monitored catalog to 441 entries.
• Expand AI-901 to the validated fundamentals depth with exact objective labels, workload/output selection, generative-model mechanics, applied responsible AI, Foundry resource/deployment/client boundaries, agent-turn diagnosis, modality-specific implementation, extraction evidence stages, and an end-to-end scenario; promote it to SOURCE-VALIDATED, register 7 exact sources, and expand the monitored catalog to 430 entries.
• Deepen the evidence layer of AI-103 across Foundry resource selection, model benchmarking/deployment, RBAC, quotas, tracing, guardrails, retrieval, agents, multimodal generation/understanding, speech, translation, and Content Understanding; promote the 8,600-word guide to SOURCE-VALIDATED, register 23 exact Microsoft sources, and expand the monitored catalog to 423 entries.
• Expand DP-900 from definitions into data-design decisions covering workload and access-pattern selection, normalization, SQL management boundaries, storage protocols, Cosmos DB partitioning/consistency/request units, analytical evidence flow, batch versus streaming, Fabric/Databricks boundaries, and Power BI modeling; promote it to SOURCE-VALIDATED, register 27 exact Microsoft sources, and expand the monitored catalog to 400 entries.
• Expand AZ-900 with an objective-by-objective decision guide covering responsibility, service models, placement, compute, request paths, storage/recovery, identity, governance, deployment, and monitoring; promote it to SOURCE-VALIDATED, register 35 exact Microsoft sources, and expand the monitored catalog to 373 entries.
• Deepen and promote GH-100 to SOURCE-VALIDATED, adding inline identity, role, ruleset, audit, App, Actions, support-bundle, and license-usage sources while retaining the deployment, GHES, migration, networking, and evidence-producing operations playbooks; register 20 exact supporting sources and expand the monitored catalog to 338 entries.
• Deepen and promote GH-500 to SOURCE-VALIDATED, adding an explicit entitlement-to-operational-coverage model and inline primary sourcing across Secret Protection, Supply Chain Security, Code Security, security operations, and suite administration; register 11 exact supporting sources and expand the monitored catalog to 318 entries.
• Deepen and promote GH-200 to SOURCE-VALIDATED, adding sourced coverage for editor validation, workflow status signals, retention and administration APIs, private templates, immutable releases, runner image dependencies, IP allow-list behavior, and secret administration; register 24 exact supporting sources and expand the monitored catalog to 307 entries.
• Promote Terraform Associate (004) to SOURCE-VALIDATED after checking all eight objective domains, 33 registered guide links, volatile HCP Terraform claims, learning-resource access and timing, and exam-integrity boundaries against the unchanged Terraform 1.12 blueprint.
• Add HashiCorp as the first non-Microsoft-platform provider and publish a 5,800-word Terraform Associate (004) draft mapped to all eight official domains, with six hands-on labs, 004-specific related depth, and individually reviewed learning options.
• Replace hard-coded GitHub/Microsoft website loops with provider records from data/vendors.json, support honest unweighted objective maps, and add a tested HashiCorp Developer objective-monitor adapter.
• Register 33 Terraform blueprint, product-documentation, official-training, course, video, and assessment sources, expanding the monitored source catalog to 283 entries.
• Add free Microsoft Practice Assessments for all 13 guides, identify the seven Pluralsight paths that explicitly include practice exams, link O'Reilly's five relevant interactive practice tests, and add the Whizlabs AB-900 course/practice offering; the monitored source catalog now has 250 entries.
• Clarify the recommended assessment sequence: start with Microsoft's free readiness check, then add a paid bank only when a different question style or explanation set closes a specific gap.
• Add 11 exam-specific MeasureUp practice tests and four additional Whizlabs training/assessment packages, expanding the monitored source catalog from 219 to 234 entries.
• Add practice-assessment selection and usage guidance, realistic review-time estimates, exam-integrity boundaries, and explicit caveats for stale or contradictory provider pages.

0.1.1 — 2026-08-31

• Suppress environment-dependent YouTube title, canonical-URL, and duration comparisons while retaining reachability and redirect monitoring.
• Verify the weekly source-health workflow through the protected pull-request path.

0.1.0 — 2026-08-31

Initial public-library release:

• publish a searchable, responsive GitHub Pages site for 13 GitHub and Microsoft certification guides;
• add generated guide-start panels with prerequisites, reading estimates, weighted objectives, study paths, first-lab links, and correction links;
• promote GH-900 and GH-300 to SOURCE-VALIDATED with machine-readable review evidence;
• register 219 exact blueprint, product-documentation, training, and expert-resource sources;
• add weekly official-objective and source-health monitoring that creates review work without rewriting content automatically;
• add structured issue forms, security-reporting guidance, social preview metadata, JSON-LD, robots and sitemap support, and accessibility evidence; and
• document GitHub Pages publication and downstream work-mirror behavior.