Quantum Ads MCP
Casius999/quantum-ads-mcp
Plano de controle MCP para Google marketing stack com 20 conectores integrados
Ver no GitHubPlano de controle soberano para o stack de marketing do Google via Model Context Protocol. Oferece 20 conectores que abrangem produtos como Google Ads, GA4, GTM, YouTube e Merchant Center, com suporte a leitura e escrita segura, auditoria assinada em Ed25519 e autenticação OAuth 2.1 multi-tenant.
Instalação
npx convez-ai Casius999/quantum-ads-mcp1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Automatizar campanhas de Google Ads via agentes de IA
- Integrar dados de GA4, GTM e outros produtos Google em fluxos de trabalho
- Gerenciar múltiplas contas de marketing com segurança e auditoria
Veredito

Implementação técnicamente sólida e completa do protocolo MCP para o stack Google, com 99% de cobertura de testes e rigor em segurança (auditoria assinada, OAuth 2.1). Idealmente para equipes de marketing e agências que precisam orquestração por IA, mas ainda em hardening para produção (aguarda transporte remoto OAuth e observabilidade).
README
A tradução para o português deste README ainda está na fila.
quantum-ads-mcp
The most complete open Model Context Protocol control plane for the Google marketing stack — 20 connectors, read + guarded write, Ed25519 signed audit, OAuth 2.1 multi-tenant.
Architecture · The 20 connectors · Install · Configure · Safety · Roadmap
A sovereign Google marketing-agency control plane over the Model Context Protocol: a shared core (multi-tenant auth, version resilience, a GAQL/query engine, a safety spine with a signed audit ledger) and 20 pluggable per-product connectors spanning the entire Google marketing stack. Built for an AI agent to operate that stack at full granularity, safely.
Status. The full connector surface is built and tested at the logic layer: read + guarded write across 20 products, every tool exercised by mocked tests (no live SDK needed). The per-product live SDK glue is isolated in
sdk.pymodules and smoke-gated (verified against real APIs with your own credentials, not in CI). Remaining for full production hardening: remote OAuth 2.1 transport + multi-tenant secret store + observability (see Roadmap). Read-only by default. Every "most X" claim here is meant to be verifiable.
Contents
Why · The 20 connectors · Architecture · Core · Install · Configure · Run · Safety & honest claims · Roadmap
Architecture
flowchart TD
Client["MCP client (AI agent)"] -->|"stdio · or OAuth 2.1 HTTP (multi-tenant)"| Server["FastMCP server"]
Server --> Core["Sovereign core: auth · version resilience · GAQL/query engine · safety spine (read-only default + Ed25519 signed audit) · registry / meta-tool"]
Core --> C1["google_ads · ga4 · gtm · merchant"]
Core --> C2["datamanager · searchconsole · youtube"]
Core --> C3["dv360 · cm360 · sa360"]
Core --> C4["bigquery · adh · looker · meridian"]
Core --> C5["vertex · trends · gbp · language · workspace · recaptcha"]
C1 -. "smoke-gated sdk.py" .-> G["Google APIs (Ads v24, GA4, GTM, Merchant, …)"]
C4 -. "smoke-gated sdk.py" .-> G
Why this exists
Surveying June 2026: Google's official MCP is read-only (3 tools); the best open write-capable servers are pinned to sunsetting API versions, untested, single-product, or have no safety layer; the capable commercial ones are closed and remote-only. No open, self-hostable server is simultaneously complete across the stack, write-capable, safe, tested, version-resilient, and multi-product. Quantum ADS fills that gap.
The 20 connectors
| Domain | Connectors |
|---|---|
| Paid media | Google Ads (v24), DV360, CM360, SA360 |
| Measurement | GA4 (Data + Admin), Data Manager API (Customer Match + conversions), Search Console |
| Catalog / tagging | Merchant API, Google Tag Manager (+ server-side, Consent Mode v2) |
| Channels | YouTube (Data + Analytics + Reporting), Business Profile, Trends |
| Warehouse / BI / MMM | BigQuery (cost-guarded), Ads Data Hub, Looker, Meridian (Bayesian MMM) |
| Creative / ops | Vertex AI (Gemini / Imagen 4 / Veo 3), Cloud Translation + NL, Workspace (Sheets/Drive/Slides), reCAPTCHA Enterprise |
Full tool catalog, backend keys, and required SDK packages: CONNECTORS.md.
Core (shared by every connector)
- Multi-tenant auth + secret store — credentials load from env / secret manager and never reach the LLM.
- Version resilience — pinned v24, sunset guard (days-until-sunset), unknown-enum tolerance (survives Google's monthly API releases and the 2026-06-10 DV360 Demand Gen rollout).
- Query engine — GAQL build / validate (single-FROM, no-
OR, segment rules, 37-month cap) / stream / flatten, with quota token-buckets + exponential backoff. SA360 uses the same query shape. - Safety spine — read-only by default; every mutation is guarded:
validate_onlypreview → two-step confirm token → Ed25519 signed audit ledger. Consent-aware (Data Manager / Consent Mode v2) and cost-aware (BigQuery dry-run, $6.25/TiB). - Connector registry + meta-tool discovery (
list_capabilities,describe_tool) and env-based connector selection (QUANTUM_ADS_CONNECTORS) so the tool list stays focused.
Install
git clone https://github.com/Casius999/quantum-ads-mcp.git
cd quantum-ads-mcp
uv sync --extra dev
uv run pytest -m "not live" # 250+ tests, no credentials needed
Per-connector live SDKs (e.g. google-analytics-data, google-api-python-client,
google-cloud-bigquery) are installed only when you wire a connector's live backend — see
CONNECTORS.md. The core + mocked test suite need none of them.
Configure
Credentials load from environment variables (no secrets on disk):
GOOGLE_ADS_USE_PROTO_PLUS=True
GOOGLE_ADS_API_VERSION=v24
GOOGLE_ADS_DEVELOPER_TOKEN=...
GOOGLE_ADS_CLIENT_ID=....apps.googleusercontent.com
GOOGLE_ADS_CLIENT_SECRET=...
GOOGLE_ADS_REFRESH_TOKEN=...
QUANTUM_ADS_READ_ONLY=true # writes refused until set to false
QUANTUM_ADS_CONNECTORS=google_ads,ga4,gtm # optional: mount a subset (default: all 19)
Never commit
.env,google-ads.yaml, orclient_secret*.json— all gitignored, andgitleaksruns in CI. See SECURITY.md.
Run
python -m quantum_ads # stdio MCP server
Wire it into your MCP client (Claude Desktop / VS Code) as a stdio server.
Safety & honest claims
Read-only by default; tokens never reach the model; zero secrets in the repo. CI runs lint, strict type-check, the full test suite (≥90% coverage gate), CodeQL, OpenSSF Scorecard, and gitleaks.
The defensible "most powerful" claim is coverage breadth + engineering rigor + safety + version resilience, each verifiable — not "record ad performance" (that depends on budget, creative, and Google's auction, not the server). Documentation states only what the test suite proves; the per-product live SDK calls are smoke-gated and labelled as such.
Roadmap (remaining for full production hardening)
Remote Streamable HTTP + OAuth 2.1/PKCE transport · multi-tenant secret store (per-client OAuth) · observability (structured logging + tracing) · live-API conformance suite · optional anchoring of the signed audit ledger.
License
MIT — see LICENSE.
— Julien Compain · NovaQuantiX