# mureo

> Local-first, safety-gated AI ad-ops framework for Claude Code, Codex, Cursor, and Gemini. Strategy-aware agents that autonomously analyze and operate Google Ads, Meta Ads, Search Console, and GA4 — credentials never leave the operator's machine.

mureo.io is the marketing site for the open-source mureo framework
(Apache 2.0). The framework runs on the operator's own machine, makes
API calls directly to ad platforms with locally-stored credentials,
and exposes structured workflows (e.g. `/daily-check`, `/rescue`,
`/search-term-cleanup`) to AI coding agents over a CLI or MCP server.

This file is a curated entry point for AI agents and LLMs. The full
sitemap lives at https://mureo.io/sitemap-index.xml. The canonical
source code is at https://github.com/logly/mureo.

## Core

- [Home](https://mureo.io/): Product overview and positioning.
- [Manifesto](https://mureo.io/manifesto/): Why mureo exists, what it is and is not — built for agents, not for dashboards.
- [Getting started](https://mureo.io/getting-started/): Install, authenticate, and run the first command.
- [Security](https://mureo.io/security/): Threat model, credential handling, and blast-radius limits (mirrors `logly/mureo/SECURITY.md`).
- [Privacy](https://mureo.io/privacy/): Data posture for the website itself (cookieless, no tracking pixels).

## Methodology

- [Methodology overview](https://mureo.io/methodology/): How mureo decides what to flag and what to do.
- [Sample size rules](https://mureo.io/methodology/sample-size-rules/): Statistical thresholds — load-bearing constants for anomaly detection (CPA min conversions, CTR min impressions, spike/drop ratios).

## Architecture

- [Architecture](https://mureo.io/architecture/): Local-first design, agent boundaries, file layout.
- [Authentication](https://mureo.io/authentication/): Credential flow per platform — read-only vs scoped write.
- [Integrations](https://mureo.io/integrations/): Supported ad platforms and connector status.
- [CLI](https://mureo.io/cli/): Command surface and subcommand reference.
- [MCP server](https://mureo.io/mcp-server/): How to use mureo from any MCP-aware agent.
- [Strategy context](https://mureo.io/strategy-context/): The role of `STRATEGY.md` and `STATE.json` in keeping recommendations aligned with the business.

## Blog

- [How AI agents misdiagnose CPA spikes](https://mureo.io/blog/ai-agents-misdiagnose-cpa-spikes/): Why naive threshold alerts on ad cost-per-acquisition fail at the point they are most needed — and the three mechanisms mureo uses to fire only on signal.
- [The threat model of AI agents touching ad accounts](https://mureo.io/blog/threat-model-ai-agents-ad-accounts/): Where ad-ops AI agents can be exploited — prompt injection, credential exfiltration, unbounded mutations — and the four mechanisms mureo uses to contain the blast radius.
- [Strategy-driven vs metric-driven optimization](https://mureo.io/blog/strategy-driven-vs-metric-driven/): Why metric optimization alone misallocates spend, what lives in STRATEGY.md, and a worked example where two accounts with identical numbers require opposite actions.
- [Running mureo as a sidecar container](https://mureo.io/blog/running-mureo-as-a-sidecar-container/): When the local pip install is not the right fit — CI runs, agency multi-tenancy, non–Claude Code MCP clients — the new Dockerfile gives mureo a sidecar shape without giving up the local-first credential posture.
- [Anatomy of /daily-check](https://mureo.io/blog/anatomy-of-daily-check/): What mureo does when an operator types a single three-word command: four-platform data pull, strategy binding, anomaly detection, and a single correlated report.
- [Open source ad ops: an Apache 2.0 commitment](https://mureo.io/blog/apache-2-commitment/): Why mureo is open source under Apache 2.0, what stays on the operator's machine, what the project will not do, and what that means for agencies and in-house teams evaluating long-term risk.

## Source code

- [GitHub repository](https://github.com/logly/mureo): Source code, issues, and discussions (Apache 2.0).
- [PyPI package](https://pypi.org/project/mureo/): `pip install mureo`.
- [Vulnerability reporting](https://github.com/logly/mureo/security/advisories/new): Private security advisory submission.

## Optional

- [RSS feed](https://mureo.io/rss.xml): Blog updates.
- [Sitemap](https://mureo.io/sitemap-index.xml): Full URL list.
- [security.txt](https://mureo.io/.well-known/security.txt): RFC 9116 security contact.