This release adds 3 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+11 more
Affected surfaces
Summary
AI summaryBuilt-in STRATEGY.md guardrail gate adds hard‑limit enforcement for ad campaign mutations.
Full changelog
The three OSS-roadmap features merged since 0.10.17 — all strengthen mureo's control-plane value.
Added
- Built-in STRATEGY.md guardrail gate (#360): declare hard limits in a
## Guardrailssection (max_daily_budget_per_campaign,max_daily_budget_increase_pct,max_total_daily_budget,blocked_operations) and mureo refuses violating native google_ads/meta_ads mutations before dispatch, regardless of the model. Fail-open (no section ⇒ no enforcement). Official/hosted MCPs are off the data path → skills self-check as best-effort (hard enforcement tracked in #359). - Deterministic outcome evaluation
mureo_outcome_evaluate(#361): reproducible improved/regressed/inconclusive verdicts with metric directions + a noise band. Platform-agnostic — works for TikTok/plugins; daily-check uses it. - Generic anomaly detection wired for every platform (#362): the platform-agnostic
analysis_anomalies_check(median baseline from action_log; zero-spend / CPA-spike / CTR-drop) is now used for plugin / hosted / official-MCP platforms too, not just the built-in google/meta modules.
See CHANGELOG.md. PRs #360, #361, #362, #363.
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]