This release adds 7 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+11 more
Summary
AI summaryAdded seven new workflow skills covering tracking health, budget pacing, monthly reporting, experiments, audience review, ad fatigue detection, and incident postmortem.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Low |
Adds /tracking-health workflow skill for conversion-tracking audit. Adds /tracking-health workflow skill for conversion-tracking audit. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /budget-pacing workflow skill for spend vs budget pacing alerts. Adds /budget-pacing workflow skill for spend vs budget pacing alerts. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /monthly-report workflow skill for client-facing monthly digest. Adds /monthly-report workflow skill for client-facing monthly digest. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /experiment workflow skill for disciplined A/B testing. Adds /experiment workflow skill for disciplined A/B testing. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /audience-review workflow skill for persona targeting audits. Adds /audience-review workflow skill for persona targeting audits. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /ad-fatigue-check workflow skill for ad fatigue scoring and routing. Adds /ad-fatigue-check workflow skill for ad fatigue scoring and routing. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Feature | Low |
Adds /incident-postmortem workflow skill for post‑rescue timeline and root‑cause analysis. Adds /incident-postmortem workflow skill for post‑rescue timeline and root‑cause analysis. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Bugfix | Low |
Corrects stale Google Ads label from v18 to v23. Corrects stale Google Ads label from v18 to v23. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Refactor | Low |
Moves diagnostic preamble to _mureo-shared; skills now reference it. Moves diagnostic preamble to _mureo-shared; skills now reference it. Source: llm_adapter@2026-07-12 Confidence: high |
— |
| Refactor | Low |
Normalizes every bundled skill's version to the package version with parity test. Normalizes every bundled skill's version to the package version with parity test. Source: llm_adapter@2026-07-12 Confidence: high |
— |
Full changelog
Skills-focused release: 7 new workflow skills and a quality pass over the whole bundled-skill set. (Built from the v0.10.20 baseline + skills changes only.)
Added
- /tracking-health — preventive conversion-tracking audit: Meta pixel inventory +
result_indicatoraudit, Google conversion-action status, GA4 divergence flags, per-platform scorecard with ranked fixes. (#390) - /budget-pacing — month-to-date spend vs monthly budget target, run-rate and projected landing, ±10% pace alerts, Guardrails-aware daily-budget recommendations. (#390)
- /monthly-report — client-facing monthly digest: MoM comparison, Goal attainment, action-log recap with outcome verdicts, next-month recommendations. (#390)
- /experiment — designed A/B tests with evidence discipline: falsifiable hypothesis, one variable, fixed window, no peeking, per-variant
mureo_outcome_evaluateverdicts (inconclusive is a valid outcome). Meta native split tests; honest manual paths on Google Ads. (#391) - /audience-review — audits targeting against the STRATEGY.md Persona using the demographic/audience/device read tools; flags mismatches and placement waste; executes only supported mutations. (#391)
- /ad-fatigue-check — frequency (derived impressions÷reach) + CTR-decay fatigue scoring with noise guards; routes fatigued ads to /creative-refresh. (#391)
- /incident-postmortem — post-rescue timeline reconstruction, root-cause analysis, structured postmortem doc, /learn insights, preventive Guardrails proposals. No ad-platform writes. (#391)
Changed
- Skills polish (#389): the diagnostic preamble (learning insights + advisor consult) now lives once in
_mureo-shared(skills carry a pointer); every bundled skill's version is normalized to the package version (with a parity test); stale Google Ads "v18" label corrected to v23; Codex host coverage documented; weekly-report gains the diagnostic preamble.
19 operational skills ship in this release.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About logly/mureo
Framework for AI agents (Claude Code, Cursor, Codex, Gemini) to operate Google Ads, Meta Ads, and Search Console. Grounded in a local STRATEGY.md — not metric-chasing. Defense-in-depth security, local-first. Apache 2.0.
Related context
Related tools
Beta — feedback welcome: [email protected]