Amazon Ads MCP
Facilitate the management and automation of Amazon Sponsored Ads
README
Amazon Advertising MCP Server
Connect AI assistants to your Amazon Advertising data. Query campaign performance, find wasted spend, diagnose metric changes, and monitor automation rules—all through natural language.
What is this?
PPC Prophet MCP is a hosted Model Context Protocol server that gives AI assistants (Claude, ChatGPT, Cursor, etc.) direct access to your Amazon Advertising data. No local installation required—just connect and query.
This is infrastructure for AI agents, not a dashboard. It enables:
- Natural language queries against your Amazon Ads data
- Automated reporting and alerting workflows
- AI-powered PPC optimization agents
- Integration with any MCP-compatible client
Features
27+ Tools for Amazon Advertising
Performance Analytics
get_campaign_performance- Campaign metrics with flexible sorting (sales, ROAS, ACOS, spend, clicks)get_search_term_performance- Search term analysis sorted by any KPIget_asin_performance- Product-level performance metricsget_match_type_performance- Compare exact, phrase, broad, and auto targetingget_dayofweek_performance- Identify best/worst performing daysget_platform_performance- SP vs SB vs SD breakdowncompare_periods- Week-over-week, month-over-month analysis
Wasted Spend Analysis
get_wasted_spend- Find campaigns, keywords, or search terms spending without converting- Customizable thresholds and result limits
- Campaign, keyword, and search term level analysis
Root Cause Diagnosis
diagnose_change- Answer "Why did my ACOS/Sales/Spend change?"- Drill down from account → campaign → keyword level
- Factor decomposition (CPC, CVR, AOV impact)
- Bid/budget change correlation
Campaign Management
list_campaigns,get_campaign,search_campaignsupdate_campaign_status,update_campaign_budgetlist_keywords,update_keyword_bid,update_keyword_status
Rules & Automation (PPC Prophet subscribers)
list_rules- View automation rulesget_rule_details- Rule configuration and filtersget_rule_outcomes- Execution historyget_recent_rule_activity- Monitor automation across profiles
Profile Management
list_profiles,get_profile,set_current_profilelist_mcp_profiles,activate_mcp_profile,check_mcp_status
MCP Apps (Interactive Widgets)
Rich visualizations rendered directly in Claude:
- Wasted Spend Analyzer - Visual charts with sortable tables
- Campaign Performance Dashboard - Metrics overview with comparisons
- Diagnose Change - Waterfall charts showing contribution breakdown
Quick Start
1. Get Your API Token
- Sign up at ppcgpt.getppcprophet.com/register
- Connect your Amazon Advertising account
- Generate an API token at ppcgpt.getppcprophet.com/mcp
2. Configure Your MCP Client
Claude Desktop / Claude Code (Recommended)
Run this command in your terminal:
claude mcp add ppc-prophet \
--transport http \
--header "Authorization: Bearer YOUR_TOKEN_HERE" \
https://ppcgpt.getppcprophet.com/mcp
Manual Configuration
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"ppc-prophet": {
"url": "https://ppcgpt.getppcprophet.com/mcp",
"transport": "streamable-http",
"headers": {
"Authorization": "Bearer YOUR_TOKEN_HERE"
}
}
}
}
3. Start Querying
"Show me wasted spend on keywords"
"Which campaigns have the best ROAS?"
"Why did my ACOS increase this week?"
"Compare last week to the week before"
Example Queries
| Query | What It Does |
|---|---|
| "Show me top 10 campaigns by sales" | Campaign performance ranked by revenue |
| "Which keywords are losing money?" | Keywords with spend but no conversions |
| "Why did my ACOS spike?" | Root cause analysis with driver attribution |
| "Compare this month to last month" | Period-over-period metrics comparison |
| "What search terms should I add as negatives?" | Non-converting search terms |
| "Show me day of week breakdown" | Performance by Monday-Sunday |
| "Which match type performs best?" | Exact vs phrase vs broad comparison |
Technical Details
- Transport: Streamable HTTP (also supports SSE)
- Authentication: Bearer token via Authorization header
- Data Range: Last 90 days of performance data
- Response Time: Sub-100ms for most queries
Documentation
- Full Query Reference: ppcgpt.getppcprophet.com/docs/queries
- Landing Page: ppcgpt.getppcprophet.com
- MCP Setup Guide: ppcgpt.getppcprophet.com/mcp
Registry Listings
- Official MCP Registry -
io.github.philmccarty/amazon-advertising-ppc-prophet - Glama / awesome-mcp-servers
Pricing
| Plan | Profiles | Features | API Calls |
|---|---|---|---|
| Free | 1 | Read-only tools | 100/day |
| Pro ($99/mo) | 5 | Full access + write tools | Unlimited |
Support
- Documentation: ppcgpt.getppcprophet.com/docs/queries
- Email: support@getppcprophet.com
- Main Platform: app.getppcprophet.com
License
Proprietary - Requires PPC Prophet account for access.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。