A Google Ads MCP server that reads campaigns, manages budgets, pauses/enables campaigns, and adds negative keywords. It acts as the AI layer between a Google Ads account and marketing decisions, delivering structured responses via strict Pydantic models. Local deployment with your API keys is supported, and the readme excerpt emphasizes daily-ops tooling and safe data handling.
🛠️ Key Features
Read campaigns and review performance
Budget tuning and control
Pause/enable campaigns
Add negative keywords
Strict Pydantic-based responses (no raw protobuf)
🚀 Use Cases
Automate routine Google Ads management
Integrate campaign data into decision workflows
Manage budgets across multiple accounts
Groom negative keyword lists for intent refinement
⚡ Developer Benefits
Local deployment with your API keys
Open-source server suitable for production with multi-account considerations
Clear, model-driven responses for reliable agent interactions
⚠️ Limitations
Based on readme excerpt; production features may require hosting and webhook setup
Open-source deployment guidance provided; hosted SLA details may vary
The AI layer between your Google Ads account and your marketing decisions.
Battle-tested. Six tools cover the daily-ops loop — campaign listing, search-term review, budget tuning, pause/enable, and negative-keyword grooming. All responses are strict Pydantic models. No raw protobuf reaches the agent.
GadsChain Demo
☁️ Moving to production?
The open-source server runs locally with your own API keys.
For hosted infrastructure with multi-account failover, SLA guarantees,
and webhook alerts — join the managed cloud waitlist.
The Problem
Raw Google Ads API returns thousands of rows. One bad campaign structure bleeds budget silently. GadsChain reads, sanitizes, and acts on your ad data before waste compounds.
Three example prompts to send to Claude (or any MCP-compatible agent):
Use get_campaigns to show me which campaigns are bleeding budget this month
Run get_search_terms for the last 30 days and tell me which queries are wasting spend
Add "free", "cheap", "jobs" as negative keywords to campaign 12345
Run it in two terminals
bash
# Tab 1 — start the MCP server
python -m gadschain.server
bash
# Tab 2 — call a tool from a Python shell or your MCP client# Tool signatures:# get_campaigns(customer_id=None)# get_search_terms(customer_id=None, days=30, campaign_id=None)# update_budget(campaign_id, new_budget_dollars, customer_id=None)# pause_campaign(campaign_id, customer_id=None)# enable_campaign(campaign_id, customer_id=None)# add_negative_keywords(campaign_id, keywords, match_type="BROAD", customer_id=None)
How it works
Three layers between raw Google Ads output and your model:
code
Google Ads API → [Fetch] → [Transform] → [Act] → MCP Tool → AI Agent
GAQL micros→$ safe
queries enum→str mutations
CTR→% shared-budget guard
Fetch: Targeted GAQL queries — only the columns the daily-ops loop actually needs. No SELECT *, no protobuf pagination footguns.
Transform: Currency micros divided to dollars, CTR scaled to percent, enums to human strings, every nested attribute lookup tolerates missing fields without crashing.
Act: Mutations route through guard rails — REMOVED blocked on status changes, shared budgets refused (shared_budget_refused), match types validated before any mutate call. The agent never gets an exception; it gets a structured {"error": ..., "message": ...} it can reason about.
Real numbers from a live Franka Pizzeria account (28-day window)
code
RAW GOOGLE ADS PAYLOAD GADSCHAIN OUTPUT
─────────────────────────────────────────────────
Impressions: 3,389 Spend (28d): $51.41
Clicks: 163 Conversions: 3 ($17.14 each)
CTR: 4.81% Conv. rate: 1.84%
Cost/click: $0.32 avg Surface: Display Network waste
identified on Fridays
($0.11 CPC vs $0.44 avg)
In one read of a real account, GadsChain surfaced $51.41 spent over 28 days for 3 conversions at $17.14 each — a 1.84% conversion rate hidden inside a 4.81% CTR that looks healthy on paper. The Display Network was the silent culprit, with Friday clicks averaging $0.11 CPC vs the $0.44 search-side average — cheap junk traffic inflating CTR while contributing nothing to conversions. The agent saw it because the transformed payload made channel attribution legible instead of buried in protobuf.