What's new in The Google Ads AI Agentic System
Every version since launch — features, content updates, fixes, and pricing changes.
v1.2.0
3
2w ago
25 days
April 2026
2026-04 Operating-System Refresh — 114 Commands, Page Review Panel, 5 Autogrowth Loops
End-to-End Codebase jumps 98 → 114 commands. Agent Codebase jumps 27 → 34 commands. Module 22 added covering Phase 0 briefs, Page Review Panel, the five autogrowth loops, PMax + Search safety defaults, Canvas-design default, and Codex compatibility.
Summary
End-to-End Codebase jumps 98 → 114 commands. Agent Codebase jumps 27 → 34 commands. Module 22 added covering Phase 0 briefs, Page Review Panel, the five autogrowth loops, PMax + Search safety defaults, Canvas-design default, and Codex compatibility.
Highlights
- Agent Codebase — 27 → 34 commands (+7 net-new)
- End-to-End Codebase — 98 → 114 commands (+16 net-new)
- Page Review Panel — Scanner + expert reviewer scoring gates every page deploy; feeds attribution + brand calibration
- Five autogrowth loops — presell-autogrow, offer-autogrow, search-autogrow, pmax-autogrow, demandgen-autogrow — all ledger-tracked, dry-run default
- Canvas-design default — Saved themes opt-in only; from-scratch design on every new build
- Module 22 added — 2026-04 operating-system update: Phase 0 briefs, PMax + Search safety defaults, AI image/video framework refresh, Codex compatibility
- 114-Command Reference Manual — Bonus renamed from 98-Command → 114-Command; Command Center Agent Spec expanded
Full release notes
Scope: End-to-End Codebase 98 → 114 commands. Agent Codebase 27 → 34 commands. Module 21 (2026-04 Doctrine Refresh) extended; Module 22 (2026-04 Operating-System Update) added.
Net-new commands (End-to-End, +16): presell-review, offer-review, presell-review-backfill, presell-smoke-test, presell-rollback, search-mine, search-autogrow, spy-refresh, pmax-autogrow, demandgen-autogrow, seasonal-creative-refresh, spy-demand-sidecars, creative-refresh-review, brand-voc-mine, brand-calibrate (extended with expert reviewer weighting), and page-health signal integration into attribution.
Net-new commands (Agent Codebase, +7): presell-review, offer-review, presell-review-backfill, presell-smoke-test, presell-rollback, search-mine, search-autogrow, pmax-autogrow, brand-voc-mine, creative-refresh-review.
Curriculum updates: Bonus 1 renamed 98-Command Reference Manual → 114-Command Reference Manual. Cross-tier comparison table now includes Page Review Panel, Canvas-design default, and the five autogrowth loops. Tier descriptions explicitly call out 2026-04 deltas.
Global framework changes (apply across every product)
- Phase 0 objective brief required on presell, offer, Search, PMax, Demand Gen, and review artifacts before generation begins
- Page Review Panel where relevant — scanner + expert reviewer scoring, P0/P1/P2 findings, combined page-health score, trust-level gating
- PMax safety defaults — URL expansion always OFF, dry-run mutations, explicit
--applyrequired - Search framework rules — exact-match only, CPC floor/ceiling, IF-function ban, single DKI per RSA, headline composition standards
- Canvas-design default — saved themes opt-in only via
theme=<name>or explicit--auto-matchflag - AI image/video framework refresh — campaign brief fields, DR creative standards, creative feedback loops, asset posteriors, multi-aspect video standards
- AutoResearch eval system — review eval buckets added for
presell-reviewandoffer-review - Claude Code + Codex compatibility — CLAUDE.md remains the Claude Code entrypoint; AGENTS.md symlinked for Codex
Security / client-readiness
- No API keys, OAuth secrets, service-account files,
.envfiles, or runtime state files shipped in any codebase - No source-brand or client names included
- No maintainer-local paths included
- MCP configs use placeholders or environment variables only
- READMEs reflect current command counts and current 2026-04 product positioning