All articles
Tutorial 22 min read

Run Google Ads and Meta Campaigns from Claude Code: The Complete Setup

A

Adspirer Team

Share Y
Run Google Ads and Meta Campaigns from Claude Code: The Complete Setup
Summary

This is the hands-on guide for running Google Ads and Meta campaigns from Claude Code. You’ll install the Adspirer plugin (one command), create your brand workspace with CLAUDE.md and STRATEGY.md, understand the 5 ad management skills, configure explicit safety rules, and run your first autonomous performance audit across both platforms. Setup takes about 10 minutes. What you get is a fully agentic workflow — scheduled audits, automated waste detection, multi-step campaign builds — all driven by Claude Code calling Adspirer’s 200+ advertising tools.

If you’ve used Claude Code for software development, you already know how it operates: it reads files, runs commands, takes multi-step actions, and can be scheduled to work unsupervised. This guide applies that same capability to paid media management — specifically Google Ads and Meta Ads.

Adspirer is a hosted MCP server that gives Claude Code access to 200+ advertising tools spanning Google Ads, Meta Ads, LinkedIn Ads, and TikTok Ads. When you connect the two, Claude Code becomes a fully autonomous ad agent. It can run a weekly performance audit every Monday morning without you opening a dashboard, find wasted spend on demand, and build complete campaigns from a one-paragraph brief — all from your terminal.

This guide focuses on the Google Ads + Meta Ads workflow. It’s more setup than connecting ChatGPT to the Adspirer plugin — but it’s meaningfully more powerful for automation. You define the rules once; Claude Code executes them on a schedule.

Info

This guide is part of the Claude Code agent setup series.

Prerequisites

Before starting, make sure you have all four of the following:

  • Claude Code CLI installed and working (npm i -g @anthropic-ai/claude-code, requires Node.js 18+). See the Claude Code installation guide.
  • Claude Pro, Max, Team, or Enterprise subscription — Claude Code requires an Anthropic paid plan.
  • Adspirer account with at least one ad account connected. Create a free account →
  • Google Ads and/or Meta Ads account connected inside Adspirer. You’ll do this during Step 2.
Tip

You can use this guide with Google Ads only, Meta Ads only, or both connected simultaneously. The cross-platform comparison features require at least one account on each platform.

Step 1: Install the Adspirer Plugin

Open Claude Code and type this directly into the input bar (not as a prompt to the AI):

/plugin install adspirer-ads-agent

Or browse for it: type /plugin, switch to the Discover tab, search for “Ads”, and select adspirer-ads-agent from claude-plugins-official.

After installation, reload plugins to activate:

/reload-plugins

This single install sets up everything needed to manage ads from Claude Code:

ComponentLocationPurpose
MCP ServerPlugin configConnects Claude Code to Adspirer’s 200+ advertising tools via OAuth 2.1
Sub-Agent~/.claude/agents/performance-marketing-agent.mdDefines the agent’s behavior, brand context loading, and workflow patterns
Skill~/.claude/skills/ad-campaign-management/SKILL.mdProven tool sequences for each task type across all platforms
5 Slash Commands~/.claude/commands/*.mdQuick-access commands for setup, performance review, waste audit, copy writing, and refresh
Warning

The /plugin and /reload-plugins commands must be typed directly into Claude Code’s input bar — they are user commands, not prompts for the AI. If you ask the AI to run them, you’ll get a “nested session” error.

Prefer manual setup without the plugin?

If you want to add just the MCP server connection without the plugin:

Terminal window
claude mcp add --transport http adspirer https://mcp.adspirer.com/mcp

To also install the sub-agent, skill, and slash commands manually:

Terminal window
git clone https://github.com/amekala/ads-mcp.git /tmp/ads-mcp
mkdir -p ~/.claude/agents ~/.claude/skills ~/.claude/commands
cp /tmp/ads-mcp/plugins/claude-code/adspirer/agents/performance-marketing-agent.md ~/.claude/agents/
cp -r /tmp/ads-mcp/plugins/claude-code/adspirer/skills/ad-campaign-management ~/.claude/skills/
cp /tmp/ads-mcp/plugins/claude-code/adspirer/commands/*.md ~/.claude/commands/

See the Claude Code setup documentation for details.

Step 2: Connect Your Ad Accounts

After installing the plugin, authenticate with Adspirer. Type in the Claude Code input bar:

/mcp

Find plugin:adspirer-ads-agent:adspirer MCP Server in the list and select Authenticate. A browser window opens where you:

  1. Sign in to your Adspirer account (or create one)
  2. Connect your Google Ads account via OAuth
  3. Connect your Meta Ads account via OAuth

Both connections use OAuth 2.1 with PKCE. Your ad credentials are never stored locally — Adspirer handles all platform authentication server-side. You can revoke access anytime from your Adspirer dashboard. See the security documentation for the full security model.

After authenticating, run /mcp again to verify. The Adspirer server should show as connected with tools listed.

Step 3: Create Your Brand Workspace

Navigate to the folder where you keep your brand’s marketing files, then launch Claude Code:

Terminal window
cd ~/Clients/YourBrand    # wherever your brand docs live
claude

Run the setup command:

/adspirer:setup

What Happens During Setup

Scans Your Local Files

The agent reads brand documents in your folder — .md, .txt, .csv, .pdf, .yaml, .json files. Brand guidelines, media plans, audience research, competitor notes — anything it finds gets absorbed into the brand context. Drop your brand docs into the folder before running setup. The more context you give it, the more relevant every future recommendation will be.

Pulls Your Ad Account Data

Calls Adspirer’s get_connections_status tool to confirm which platforms are live, then pulls account-level metadata for Google Ads and Meta Ads — account names, IDs, active campaign counts, and current status.

Pulls 30-Day Performance Baseline

Pulls the last 30 days of campaign performance across all connected platforms. This baseline is what the agent uses for anomaly detection, KPI comparisons, and future trend analysis.

Creates CLAUDE.md

Builds the brand context file — brand overview, target audiences, voice guidelines, connected platforms, budget guardrails, KPI targets, performance snapshot, and key findings from the initial data pull. Claude Code reads this file automatically at the start of every session.

Creates STRATEGY.md

Initializes the strategy playbook. Starts nearly empty but fills over time with confirmed strategic directives: AVOID/PREFER/CONSTRAINT rules per platform — for example, “AVOID broad match on branded Google Ads campaigns” or “PREFER Reels placements on Meta for prospecting audiences.” Directives are only written to this file after your explicit confirmation.

Your Workspace After Setup

~/Clients/YourBrand/
├── CLAUDE.md           # Brand context — loaded automatically every session
├── STRATEGY.md         # Strategy directives — updated only with your approval
├── brand-guidelines.pdf  # Your existing docs
└── media-plan.csv        # Your existing docs

What a Real CLAUDE.md Contains

Here’s an abbreviated example of what the agent creates for a SaaS brand:

# AGENTS.md — CloudBase Brand Context


## Brand Overview
CloudBase sells B2B project management software for engineering teams.
Industry: SaaS / Productivity. Primary market: US, Canada, UK.


## KPI Targets
- Google Ads Search: Target CPA $80, Target ROAS 4.0x
- Meta Ads Prospecting: Target CPL $35, Frequency cap: 3.0
- Meta Ads Retargeting: Target ROAS 6.0x, Max CPM $18


## Budget Guardrails
- Monthly Google Ads budget: $12,000 (never exceed without approval)
- Monthly Meta Ads budget: $8,000 (never exceed without approval)
- Single campaign daily budget cap: $500


## Connected Platforms
- Google Ads: Account 123-456-7890 (12 active campaigns)
- Meta Ads: Account 9876543210 (8 active ad sets)


## Performance Snapshot (Last 30 Days)
- Google Ads: 847 conversions, avg CPA $74, ROAS 4.3x
- Meta Ads: 312 leads, avg CPL $41, top creative: Video-ProductDemo-v3


## Brand Voice
Professional but approachable. No jargon. Short sentences. Never use
"leverage", "synergy", or "game-changer". Approved CTA: "Start free trial".


## Safety Rules
- All new campaigns created PAUSED — never enable without human approval
- Budget changes above 20% require explicit human confirmation
- Never modify campaigns that have converted in the last 7 days without review

What a Real STRATEGY.md Contains

# STRATEGY.md — CloudBase Ad Strategy


## Google Ads Directives
- AVOID: broad match on branded keywords (causes brand cannibalization)
- PREFER: exact + phrase match for high-intent keywords
- CONSTRAINT: max CPC $4.00 on non-branded search terms


## Meta Ads Directives
- PREFER: Advantage+ audience over manual interests for prospecting
- AVOID: static image creatives in prospecting (video outperforms 2.4x)
- CONSTRAINT: pause any ad set with frequency > 3.0 within 7 days


## Cross-Platform
- Attribution model: Data-driven (Google), 7-day click / 1-day view (Meta)
- Report comparison window: same 30-day period, not rolling

Step 4: The 5 Ad Management Skills

Your agent has 5 skills covering every ad management workflow. You don’t need to memorize slash commands — describe what you need in plain language and the agent matches the right workflow. But knowing the commands makes you faster when you know exactly what you want.

SkillSlash CommandWhat It Does
Setup/adspirer:setupBootstraps brand workspace, creates CLAUDE.md and STRATEGY.md, pulls account data
Performance Review/adspirer:performance-reviewCross-platform scorecard against your KPI targets, anomaly detection, strategy drift check
Wasted Spend/adspirer:wasted-spendFinds keywords spending without converting, creatives with declining CTR, audiences with poor ROI
Ad Copy/adspirer:write-ad-copyBrand-voice ad copy using your CLAUDE.md guidelines and real performance data
Refresh/adspirer:refresh-brand-contextRe-scans folder for new brand docs, pulls latest performance, updates CLAUDE.md

Skill: /adspirer:performance-review

Runs a cross-platform scorecard across all connected platforms. Pulls campaign-level metrics, compares against your KPI targets from CLAUDE.md, checks STRATEGY.md for strategy drift (campaigns violating your directives), and identifies top performers, underperformers, and anomalies.

For Google Ads this includes campaign metrics, keyword performance, search term reports, and Quality Score trends. For Meta Ads this includes ad set performance, creative fatigue signals, frequency data, and audience overlap indicators.

Cross-Platform Performance Review

Run a performance review across all my connected platforms for the last 30 days. Compare each campaign to my KPI targets from CLAUDE.md. Flag any campaigns with CPA above target, declining ROAS, or creative fatigue. Check STRATEGY.md for any strategy drift. Give me the top 3 actions I should take today.

Skill: /adspirer:wasted-spend

Finds money leaks across Google Ads and Meta Ads. For Google Ads: keywords spending without converting, search terms triggering irrelevant queries, ad groups with poor CTR dragging down Quality Scores. For Meta Ads: creatives with declining performance past their effective lifespan, audiences with high frequency and low conversion rates.

Returns a prioritized list of cuts with estimated monthly savings for each.

Weekly Waste Audit

Find wasted spend across all my ad platforms for the last 14 days. For Google Ads, show keywords that spent over $50 without a single conversion. For Meta Ads, show any creatives with CTR below 0.8% or frequency above 3.0. Total up potential monthly savings.

Skill: Natural language campaign creation (uses the ad-campaign-management skill)

Build complete campaigns from a brief. The agent researches keywords (Google Ads) or audience segments (Meta Ads), validates your brand voice and budget guardrails against CLAUDE.md, creates the campaign structure, and then stops — all campaigns are created PAUSED. Nothing goes live until you review and manually enable.

Campaign Creation from a Brief

Create a Google Search campaign for our new enterprise tier launch. Budget $150/day. Research high-intent keywords for “project management software enterprise”. Create 3 ad groups by intent cluster with RSAs using my brand voice from CLAUDE.md. Add sitelinks for the enterprise features page, pricing page, and case studies.

Skill: /adspirer:write-ad-copy

Generates brand-voice ad copy using your CLAUDE.md guidelines and real performance data from your account. The agent knows which headlines and descriptions have performed best in your account, what your brand prohibitions are, and what character limits each platform and format requires.

For Meta Ads, it generates primary text, headlines, and descriptions in multiple variants for testing. For Google Ads, it generates RSA-optimized headlines and descriptions with pin recommendations.

Meta Ads Creative Brief

Write Meta Ads copy for a retargeting campaign targeting website visitors who viewed our pricing page but didn’t convert. Use my brand voice from CLAUDE.md. Generate 3 variants of primary text, 5 headline options, and 3 description variants. Reference what’s worked in my account and flag any copy that’s too similar to current live ads.

Skill: /adspirer:refresh-brand-context

Re-scans your project folder for new brand documents and pulls the latest performance data from all connected platforms. Updates CLAUDE.md with current metrics and any new brand context. Run this after connecting new ad accounts, adding new brand docs, or when you want to refresh the performance baseline after a campaign launch.

Refresh After Campaign Launch

Refresh my brand context — I launched a new Meta Ads campaign last week and connected a second Google Ads account. Pull the latest data for all accounts and update CLAUDE.md with the new performance baseline.

Step 5: Configure Safety Rules

This is the most important section of this guide. When Claude Code manages live ad spend autonomously, the safety rules you define in CLAUDE.md are your protection layer. They are not suggestions — they are constraints the agent enforces before taking any action.

Add this safety block to your CLAUDE.md before you run any write operations:

## Safety Rules (Non-Negotiable)


### What the Agent MUST NEVER Do Without Explicit Human Approval
- Enable any campaign from PAUSED to ENABLED status
- Increase any campaign's daily or monthly budget
- Change any campaign's bidding strategy
- Modify a campaign that has converted in the last 7 days
- Create campaigns on platforms not listed in Connected Platforms above


### What the Agent Can Do Autonomously (Read Operations)
- Pull performance data from any connected platform
- Generate reports, audits, and recommendations
- Identify wasted spend candidates
- Research keywords and audience segments
- Generate ad copy drafts


### Spending Caps — Hard Limits
- No single campaign may have a daily budget above $500 without approval
- No monthly platform budget may exceed the amounts in KPI Targets above
- Any budget increase recommendation above 20% of current spend requires human sign-off


### Campaign Creation Rules
- All new campaigns created with status: PAUSED
- Include a summary of all campaigns created in the same response
- Flag any campaign where estimated CPC exceeds the CPA target by 3x
Warning

Never enable Claude Code’s “Full Access” mode for ad management workflows. Full Access auto-approves all tool calls including write operations. For advertising, always use Auto mode — read operations run automatically, write operations (creating campaigns, changing budgets, modifying targeting) require your explicit approval before executing. The whole value of the safety model is that the agent surfaces the recommendation and you make the final call on spend.

Claude Code’s Built-In Permission Layer

Beyond your CLAUDE.md safety rules, Claude Code enforces a second layer through its approval modes:

ModeTool CallsRight For
DefaultEach requires approvalFirst-time setup and learning the system
Auto (recommended)Read ops auto-approved, write ops prompt youDay-to-day ad management
Full AccessAll auto-approvedNever use for ad management

In Auto mode, when the agent tries to create a campaign, Claude Code surfaces the exact API call it’s about to make and waits for you to approve or reject. This is not a speed bottleneck — it’s the natural check that experienced media buyers would want before any spend commitment.

Step 6: Run Your First Performance Audit

With installation, authentication, brand workspace, and safety rules in place, run your first audit:

First Audit — Google Ads and Meta

Run a performance review across my Google Ads and Meta Ads accounts for the last 30 days. For Google Ads: show campaign-level metrics, flag any keywords with CPA above $120, and check the search terms report for obvious negatives. For Meta Ads: show ad set performance, flag any creatives with frequency above 2.5, and identify the top 3 performing audiences. Compare both platforms against my KPI targets from CLAUDE.md and give me a prioritized action list.

What Claude Code Does

After you send that prompt, Claude Code executes this sequence automatically:

Reads Your Brand Context

Loads CLAUDE.md (KPI targets, budget guardrails, brand rules), STRATEGY.md (active directives and constraints).

Verifies Platform Connections

Calls get_connections_status to confirm Google Ads and Meta Ads are active and pulls account-level metadata.

Pulls Google Ads Data

Calls Adspirer tools for campaign metrics, ad group performance, keyword data, search term report, and Quality Score indicators for the 30-day window.

Pulls Meta Ads Data

Calls Adspirer tools for campaign and ad set metrics, creative-level performance, audience segment data, frequency by ad set, and delivery insights.

Cross-References Against Targets

Compares actual CPA, ROAS, CPL, and frequency against your CLAUDE.md targets. Checks STRATEGY.md for any campaigns violating your directives — for example, a broad match keyword running on a branded campaign you explicitly prohibited.

Returns Structured Report

Delivers a prioritized action list — metrics by platform, anomalies flagged with explanations, strategy drift identified, and specific recommended actions ordered by estimated impact.

What the Output Looks Like

The agent returns a structured report covering:

  • Platform scorecards — campaign-by-campaign comparison against your KPI targets, color-coded by performance vs. target
  • Anomalies — any metrics outside normal range with a causal explanation (e.g., “Search Campaign #3 CPA spiked 40% — driven by 3 keywords with zero conversions in 30 days and $800 combined spend”)
  • Strategy drift — any campaigns running in violation of your STRATEGY.md directives
  • Wasted spend candidates — specific line items with dollar amounts and recommended action
  • Top 3 actions — prioritized by estimated impact with the data reasoning for each

No dashboard tabs. No CSV exports. No spreadsheet building. From terminal prompt to structured analysis in a few minutes.

What Autonomous Mode Actually Means

“Autonomous” in Claude Code means something specific — and something more limited than the word might imply. It’s worth being direct about this.

What Claude Code Can Run Unsupervised

These are read-only operations. Claude Code can be scheduled to run these on a cron schedule without your presence:

  • Pull performance data and generate reports
  • Flag anomalies against your KPI targets
  • Identify wasted spend candidates
  • Check for strategy drift
  • Generate copy drafts
  • Research keywords and audience segments

What Always Requires Your Approval

These are write operations that affect live campaigns and real spend. Claude Code will stop and wait for you before executing any of these, regardless of your settings:

  • Enabling any campaign from PAUSED to LIVE
  • Changing any budget (daily or monthly)
  • Modifying targeting on a live ad set
  • Adding or removing keywords
  • Changing bidding strategies
Warning

The model where an AI “runs your ads without you” is the wrong frame. The right frame: the agent handles the analysis, research, and drafting — the parts that consume 80% of the time but require 20% of the judgment. You make every call that involves live spend. This is the boundary that makes autonomous ad management practical rather than reckless.

Setting Up Scheduled Audits

To schedule a weekly performance check using Claude Code’s scheduling system:

Terminal window
claude schedule create \
  --name "Weekly Google and Meta Audit" \
  --cron "0 8 * * MON" \
  --prompt "Run a performance review across Google Ads and Meta Ads for the last 7 days. Compare against KPI targets in CLAUDE.md. Flag any CPA increases above 25%, any Meta creatives with frequency above 3.0, and any keywords spending over $100 without a conversion. Return a prioritized action list I can review first thing Monday morning." \
  --working-dir ~/Clients/YourBrand

This creates an agent that fires every Monday at 8 AM, reads your brand context, pulls fresh data from both platforms, and sends you a structured report — without you opening any ad interface.

Both platforms connect through the same Adspirer MCP server, but the workflows and what the agent monitors differ by platform.

What the agent tracks:

  • Campaign, ad group, and keyword performance vs. your CPA/ROAS targets
  • Search term reports — queries triggering your ads that don’t match intent
  • Quality Score trends by keyword and ad group
  • Impression share — are you losing to budget or to rank?
  • Bidding strategy performance vs. your targets

What’s different about Google Ads workflows:

  • Keyword research is a first-class step before any campaign creation
  • Search term reports are checked every audit cycle — negative keyword additions are the highest-ROI recommendation in most accounts
  • RSA ad copy is generated with explicit headline pin recommendations based on your top-converting messages
  • Smart Bidding campaigns require human approval to switch strategies — the agent flags when tROAS or tCPA targets are too aggressive

Key prompts for Google Ads:

Google Ads Search Term Audit

Pull the search terms report for all active Google Ads campaigns for the last 30 days. Filter to queries with more than $20 spend and zero conversions. Group them by pattern (brand, competitor, informational, irrelevant). Give me a negative keyword list I can review and add.

Honest limitation: Google Ads campaign creation via the API creates campaigns that require review in the Google Ads UI before enabling. The agent creates the full structure correctly — but plan to spend 5 minutes in the UI confirming ad formats and extensions look right before enabling anything.

Meta Ads from Claude Code

What the agent tracks:

  • Ad set and creative performance vs. your CPL/ROAS targets
  • Frequency by ad set — the key early warning signal for creative fatigue
  • Audience overlap between ad sets in the same campaign
  • Creative-level CTR trends over time — when is a creative past its effective lifespan?
  • Advantage+ audience performance vs. manual interest targeting

What’s different about Meta Ads workflows:

  • Creative fatigue monitoring is the highest-value recurring task — the agent flags creatives past their performance peak before you’re wasting budget
  • Audience segment analysis tells you which audiences are converting vs. just spending
  • Copy generation accounts for Meta’s primary text, headline, and description character limits across placements
  • Campaign Objective mapping matters — the agent checks that your campaign objective aligns with the conversion event you care about

Key prompts for Meta Ads:

Meta Ads Creative Fatigue Check

Check all active Meta Ads campaigns for creative fatigue signals. For each ad set with frequency above 2.5, show the CTR trend over the last 14 days. Flag any creatives where CTR has dropped more than 30% from their first-week average. For each flagged creative, suggest whether to refresh the hook, the visual, or retire the creative entirely.

Honest limitation: Meta’s API has stricter rate limits than Google Ads. For accounts with many campaigns and large date ranges, the agent may need to pull data in batches. If you get a rate limit error, break the audit into shorter time windows or fewer campaigns per request.

PromptBoxes: 5 Commands to Run This Week

Beyond the audit commands above, these are the five prompts that deliver the most value in the first week of using Claude Code with Adspirer:

Cross-Platform Performance Comparison

Compare my Google Ads and Meta Ads performance side-by-side for the last 30 days. For each platform show: total spend, conversions, CPA, ROAS, and % of KPI target hit. Where am I underperforming on each platform and what’s the most likely cause? Which platform should I put incremental budget into this month and why?

Weekly Wasted Spend Report

Find wasted spend across Google Ads and Meta Ads for the last 7 days. For Google Ads: keywords with spend above $30 and zero conversions, and search terms triggering broad match that don’t match any ad group theme. For Meta Ads: ad sets with frequency above 3.0, and creatives with CTR below 0.7%. Give me a total dollar figure I can recover and a prioritized list of what to cut first.

Campaign Creation from a Brief

Create a Meta Ads prospecting campaign targeting US-based B2B SaaS decision-makers. Budget $200/day. Use Advantage+ audience targeting. Create 2 ad sets: one for video creative and one for static image. Use my brand voice from CLAUDE.md for all copy. Create all campaigns and ad sets with status PAUSED — I’ll review and enable after you summarize the structure.

Scheduled Monitoring Setup

Set up a recurring Monday morning audit that checks both Google Ads and Meta Ads for the prior week. The report should flag: any campaign with CPA more than 20% above my target, any Meta creative with frequency above 3.0, any Google Ads keyword with zero conversions and more than $50 spend, and any budget pacing issues. I want the report waiting for me when I open my terminal Monday morning.

FAQ

Does Claude Code work as an MCP client for Adspirer? How is this different from connecting Claude Desktop?

Yes — Claude Code connects to Adspirer via the same MCP protocol as Claude Desktop. The key difference is Claude Code’s agentic capabilities: it can be scheduled to run tasks, execute multi-step workflows autonomously, read and write local files (your CLAUDE.md and STRATEGY.md), and is invoked from the terminal rather than a chat interface. Claude Desktop is better for conversational ad analysis. Claude Code is better for repeatable autonomous workflows.

Which Adspirer pricing tier do I need for regular Google Ads and Meta management?

The Free tier gives you 15 tool calls/month — enough to try one performance review. For regular use, the Plus tier (150 calls/month) covers weekly audits on 2 platforms plus occasional campaign creation. Pro (600 calls/month) covers daily monitoring, weekly deep audits, and active campaign builds. A cross-platform review uses 4-8 tool calls depending on account size and how many campaigns are active. See Adspirer pricing for current tiers.

Is it safe to connect both my Google Ads and Meta Ads accounts to Adspirer?

Both connections use OAuth 2.1 — the same protocol your CRM or reporting tools use to connect ad platforms. Adspirer requests read and write permissions (write is needed for campaign creation and optimization). Your credentials are stored server-side encrypted, never locally. The MCP tools are scoped — the agent can only do what the Adspirer API exposes, and every write operation is logged in your Adspirer dashboard. You can revoke access for either platform independently at any time. See the security documentation for the full scope list.

Can I use this with just Google Ads if I don't run Meta campaigns?

Yes. The plugin and all skills work with a single connected platform. If you only have Google Ads connected, the performance review, wasted spend, and campaign creation skills all work fully for Google Ads only. Cross-platform comparison features gracefully skip platforms with no active connection. You can connect Meta Ads later and run /adspirer:refresh-brand-context to pull the new account into your workspace.

How is this different from just using Claude.ai or ChatGPT with Adspirer's MCP connector?

Three meaningful differences for automation use cases. First, Claude Code can be scheduled — you can set up a Monday morning audit that runs without you opening any interface. Claude.ai and ChatGPT require a human to initiate each session. Second, Claude Code reads and writes files on your machine — your CLAUDE.md and STRATEGY.md persist the brand context that makes every session smarter than the last. Third, Claude Code’s sub-agent architecture runs complex workflows in an isolated context window, meaning a 30-campaign cross-platform audit doesn’t flood your conversation history. For one-off analysis, ChatGPT or Claude.ai with the Adspirer MCP connector is simpler to set up. For repeatable autonomous workflows, Claude Code is more powerful.

What’s Next

Your setup is complete: the Adspirer plugin is installed, Google Ads and Meta Ads are connected, your brand workspace is created, safety rules are configured, and your first performance audit is done.

The two highest-value next steps from here:

  1. Set up the weekly scheduled audit — the Monday morning prompt above. Once this runs automatically, you stop missing the Friday CPA spike that compounds over the weekend.
  2. Build out your STRATEGY.md — after each audit session where the agent identifies a pattern (e.g., “broad match on branded terms is cannibalizing conversions”), add a confirmed CONSTRAINT directive. Over time, this file becomes the accumulated knowledge of every insight you’ve validated.

The compounding effect is real: the more sessions you run, the more context the agent carries into the next one. Three months in, Claude Code is not starting fresh every Monday — it’s working from a complete picture of your account history, your brand constraints, and your proven strategies.

Info

Ready to start? Sign up for Adspirer (free tier: 15 tool calls/month) and install the plugin with /plugin install adspirer-ads-agent.

Get started free →

Claude Code Google Ads Meta Ads AI Agents MCP Automation Tutorial

More articles to read

Best AI Ad Management Platform 2026: Complete Buyer's Guide
Deep Dive

Best AI Ad Management Platform 2026: Complete Buyer's Guide

7 AI ad management platforms compared honestly — including the new category of MCP-based tools that work inside Claude and ChatGPT. Who should use what, real pricing, and the trade-offs nobody else explains.

A
Adspirer Team ·
Should You Hire a PPC Specialist or Use AI? 2026 Guide
Deep Dive

Should You Hire a PPC Specialist or Use AI? 2026 Guide

Honest comparison of hiring a PPC manager vs using AI tools like Claude and ChatGPT to run Google Ads and Meta campaigns in-house. Includes real cost breakdown.

A
Adspirer Team ·