A senior AI media buyer at your fingertips. Meta strategists, Google Ads specialists, creative-fatigue analysts, GA4 attribution experts, Shopify ops leads — each a specialized agent with battle-tested processes, not a generic chatbot.

AI Media Buyer is the AI media buyer for Claude Code & Codex CLI — it plugs your terminal into your ad accounts and turns it into a senior media buyer across Google Ads, Meta, TikTok, LinkedIn, Bing, GA4, Shopify, Klaviyo & Search Console. Built by GoMarble, the team that runs paid media for some of the fastest-growing DTC brands.

• 🔌 Live ad-platform data via the GoMarble MCP server — OAuth, one-click sign-in
• 🧠 36 expert-authored skills — Google Ads, Meta (Facebook/Instagram), GA4, Shopify, Search Console, Klaviyo, TikTok, LinkedIn, Bing Ads, document generation
• ⚡ 8 morning-workflow slash commands — daily audits and decision matrices for Meta + Google
• 🛡️ Built-in guardrails — no synthetic data, no fabricated keywords, proper attribution discipline

Install

Claude Code

In a Claude Code session, paste these three commands:

/plugin marketplace add https://github.com/gomarble-ai/ai-media-buyer.git
/plugin install ai-media-buyer@ai-media-buyer
/reload-plugins

Using the full HTTPS URL (not the owner/repo shorthand) avoids SSH-key errors on machines where git is configured to rewrite GitHub HTTPS URLs to SSH.

Then connect the MCP server:

/mcp

Pick gomarble, click Authenticate — the browser opens, you sign in to GoMarble, you're done.

Codex CLI

codex plugin marketplace add https://github.com/gomarble-ai/ai-media-buyer.git

Start Codex, open the Plugins panel, find AI Media Buyer, and install. Then:

codex mcp login gomarble

Same OAuth flow, one-time.

One-command install for both (via npm)

npx ai-media-buyer

This wraps the official commands above and works on macOS / Linux / Windows.

🗣️ What you can ask the agent

The skills auto-invoke based on what you ask — you don't have to remember anything. Some examples:

| Ask | What happens | |---|---| | "Audit my Meta account act_12345 for the last 30 days." | Loads performance + creative + depth-of-analysis skills, pulls 30d data, runs Pareto analysis, flags fatiguing creatives, surfaces ROAS outliers, produces a report. | | "How did my Google Ads Search campaigns do this week?" | Loads search-analysis skill, classifies queries Q1–Q5, diagnoses CPC inflation or rank pressure, returns a decision matrix. | | "Create a Meta ad set for this creative targeting US iOS users." | Loads create/master + adset skills, verifies parent campaign, detects pixel, builds the propose payload, asks you to confirm before posting. | | "Pull a 7d Shopify sales report and compare to GA4 conversions." | Loads Shopify order-discipline + GA4 source-of-truth skills, queries both, reconciles the difference. | | "Generate a PPTX summarizing all my paid-media performance." | Loads pptx skill + the relevant analytics skills, produces a deck. |

⚡ Slash commands (Claude Code only)

8 read-only morning-workflow commands. Each produces analysis + recommendations — never executes mutations. Mutations happen separately via Agent Mode if the user explicitly opts in. Click any command name to view its source.

📊 Meta (4)

| Command | What it does | |---|---| | /ai-media-buyer:meta-daily-optimization <acct> | Morning briefing: 1D vs 3D vs 7D, change-log gate, root-cause action recommendations (pause / cut / reallocate / scale candidates) | | /ai-media-buyer:meta-ads-audit <acct> | 30-day comprehensive audit — pixel/CAPI, fatigue, audience split, ROAS outliers, budget allocation | | /ai-media-buyer:meta-creative-fatigue-detection <acct> | Per-ad scoring (Healthy / Early Warning / Fatigued / Dead) with refresh recommendations | | /ai-media-buyer:meta-creative-strategy <acct> | Winners + losers, pattern extraction, test plan, scaling plan, 12-creative production spec |

🎯 Google (4)

| Command | What it does | |---|---| | /ai-media-buyer:google-search-audit <acct> | Daily Search briefing: brand-vs-non-brand segmentation, CUT / FIX / SCALE recommendations | | /ai-media-buyer:google-pmax-pulse <acct> | 3d-vs-3d PMax anomaly check — Critical / Alert / Monitor classification, disciplined against overcorrection | | /ai-media-buyer:google-search-term-audit <acct> | Forensic waste audit using 80/80 Pareto + campaign-relative triggers; suggested negative keywords and root negatives | | /ai-media-buyer:google-impression-share <acct> | Lost-IS analysis → scaling opportunities (Budget) vs bid/quality work (Rank), gated on profitability |

In Codex, just describe the task in natural language — the skills' description fields handle routing.

📦 Skills shipped (36)

Click any skill name to view its SKILL.md source.

🎯 Google Ads (17)

| Type | Skill | What it does | |---|---|---| | 🏗️ Foundations | google-ads-tool-fundamentals | Tool-call patterns, GAQL basics, account structure | | 🏗️ Foundations | google-ads-guardrails | Mutation safety, attribution rules, what never to fabricate | | 🔍 Analysis | google-ads-search-analysis | Q1–Q5 query classification, CPC inflation diagnostics | | 🔍 Analysis | google-ads-shopping | Shopping / Merchant Center performance audits | | 🔍 Analysis | google-ads-pmax-evaluation | PMax 3d-vs-3d anomaly detection | | 🔍 Analysis | google-ads-pmax-scaling | PMax scaling decisions, disciplined against overcorrection | | 🔍 Analysis | google-ads-keywordplanner | Keyword discovery without fabrication | | 🔍 Analysis | google-ads-depth-of-analysis | Multi-layer drill-down methodology | | ⚡ Execution | google-ads-search-execution | Bid / budget / structure changes (gated on profitability) | | 🚀 Creation | google-ads-create-master-skill | Master orchestrator for new campaign builds | | 🚀 Creation | google-ads-create-campaign | Campaign-level setup | | 🚀 Creation | google-ads-create-ad-group | Ad-group creation | | 🚀 Creation | google-ads-create-ad | Ad-level creation | | 🚀 Creation | google-ads-create-asset | Asset uploads | | 🚀 Creation | google-ads-create-experiment | Experiment / draft setup | | 🚀 Creation | google-ads-create-negative-keyword-list | Negative-keyword list management | | 🚀 Creation | google-ads-create-bid-modifiers | Bid-modifier setup |

📊 Meta — Facebook + Instagram (10)

| Type | Skill | What it does | |---|---|---| | 🏗️ Foundations | meta-tool-fundamentals | Tool-call patterns, account / campaign / ad-set / ad taxonomy | | 🏗️ Foundations | meta-guardrails | Mutation safety, attribution discipline | | 🔍 Analysis | meta-performance-analysis | Account-level performance audits | | 🔍 Analysis | meta-creative-analysis | Creative-fatigue scoring (Healthy / Warning / Fatigued / Dead) | | 🔍 Analysis | meta-depth-of-analysis | Multi-layer drill-down for Meta accounts | | ⚙️ Operations | meta-agent-operations | Agentic-loop patterns for Meta workflows | | 🚀 Creation | meta-create-master-skill | Master orchestrator for new Meta campaign builds | | 🚀 Creation | meta-create-campaign | Campaign-level setup | | 🚀 Creation | meta-create-adset | Ad-set setup | | 🚀 Creation | meta-create-ad-with-creative | Ad + creative pairing |

📈 Other platforms (4)

| Platform | Skill | What it does | |---|---|---| | 📊 GA4 | ga4-source-of-truth | Conversions ≠ transactions, channel-subset-sum traps, attribution discipline | | 🛍️ Shopify | shopify-order-discipline | No financial_status filter, gross-vs-net, refunds, multi-currency | | 🔎 Search Console | search-console-master-skill | Organic search analysis + opportunity scoring | | 🎨 Creative | creative-research | Competitor research, evergreen + breakout winners, pattern analysis |

📄 Document generation (5)

| Format | Skill | What it does | |---|---|---| | 📝 DOCX | documents-docx-skill | Word document generation | | 🎯 PPTX | documents-pptx-skill | PowerPoint deck generation | | 📑 PDF | documents-pdf-skill | PDF report generation | | 📊 XLSX | documents-xlsx-skill | Excel spreadsheet generation | | 🐍 Python | python-sandbox-skill | Foundation for all document generation |

🔄 How it works

You ask:  "Audit my Meta account"
    ↓
Claude / Codex auto-invokes the right skills by description
    ↓
Skill instructs the agent to call GoMarble MCP tools (meta_get_account_insights, etc.)
    ↓
GoMarble MCP → live Meta Ads API call (with your OAuth token)
    ↓
Agent applies the methodology + guardrails to interpret the data
    ↓
You get the answer

The plugin ships only the methodology. All live data comes from the GoMarble MCP server you authorized with /mcp (Claude) or codex mcp login gomarble.

🔁 Updating

# Claude Code
/plugin marketplace update ai-media-buyer

# Codex CLI
codex plugin marketplace update ai-media-buyer

The SessionStart hook also surfaces an "update available" prompt automatically the first time you start a session on an outdated version.

📁 Repo layout

This is what gets installed when you run any of the install commands above:

ai-media-buyer/
├── .claude-plugin/
│   ├── plugin.json              # Claude Code manifest
│   └── marketplace.json         # marketplace declaration
├── .codex-plugin/
│   └── plugin.json              # Codex manifest with rich install metadata
├── .mcp.json                    # remote MCP wiring (Streamable HTTP + OAuth)
├── commands/                    # 8 slash commands (Claude reads; Codex ignores)
├── hooks/
│   ├── hooks.json               # registers SessionStart hook
│   └── session-start.mjs        # MCP probe + version check + today's-date injection
├── skills/                      # 36 SKILL.md folders — both hosts read from here
└── README.md

Why one folder works for both hosts

Each host looks for its manifest in a dedicated subfolder; everything else is shared.

| Host | Manifest | Reads | Ignores | |---|---|---|---| | Claude Code | .claude-plugin/plugin.json | skills/, commands/, .mcp.json (mcpServers key) | .codex-plugin/ | | Codex | .codex-plugin/plugin.json | skills/, .mcp.json (mcp_servers key) | .claude-plugin/plugin.json, commands/ |

.mcp.json ships both wrapper keys with identical content so each host finds its preferred form.

Support

• Docs & examples: https://gomarble.ai
• Issues & feature requests: https://github.com/gomarble-ai/ai-media-buyer/issues
• Account / billing: https://apps.gomarble.ai