[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_126":127},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":34,"_35":26,"_36":37,"_38":-5,"_39":-5,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":48,"_49":48,"_50":51,"_52":53,"_54":55,"_56":-5,"_57":48,"_58":59,"_60":61,"_62":-5,"_63":-5,"_64":-5,"_65":-5,"_66":67,"_68":69,"_70":71,"_72":73,"_74":75,"_76":-5,"_77":-5,"_78":-5,"_79":80,"_81":-5,"_82":44,"_83":84,"_85":86,"_87":88,"_89":90,"_91":92,"_93":94},"apiUrl","https://odel.app/api","creator","tspicer","moduleId","amazon-ads-mcp","@tspicer/amazon-ads-mcp","displayName","Amazon Ads","tagline","Amazon Ads API MCP server for Amazon Advertising API","iconKey","https://avatars.githubusercontent.com/u/231693178?v=4","category","local","status","published","visibility","public","metaDescription","deploymentType","external","workerName","bindingName","mcpConfig","pricingModel","free","priceCredits",0,"creatorUserId","system:external-import","createdAt",1790466369000,"updatedAt","sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=io.github.tspicer%2Famazon_ads_mcp","sourceVerified",true,"sourceMissingSince","importedAt","upstreamStars",70,"upstreamUpdatedAt",1788736114000,"detectedTransport","detectedAuthType","probeStatus","lastProbedAt","githubRepoUrl","https://github.com/KuudoAI/amazon_ads_mcp","githubRepoFullName","KuudoAI/amazon_ads_mcp","releaseId","ext-tspicer-amazon-ads-mcp","version","0.2.4","descriptionMarkdown","

\n\n# Amazon Ads API MCP SDK\n\n**Build AI-powered advertising applications with the Model Context Protocol (MCP) SDK for Amazon Advertising API**\n\n*Made with ❤️ + ☕ by [Openbridge](https://www.openbridge.com/)*\n\n[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![CI](https://github.com/KuudoAI/amazon_ads_mcp/actions/workflows/ci.yml/badge.svg)](https://github.com/KuudoAI/amazon_ads_mcp/actions/workflows/ci.yml)\n\n**MCP registry id** (for clients and catalogs that display a stable package name): `io.github.KuudoAI/amazon_ads_mcp`\n\n

\n\n## Table of contents\n\n- [Where this fits in Kuudo](#where-this-fits-in-kuudo)\n- [What are MCP tools?](#what-are-mcp-tools)\n- [What is Amazon Ads API MCP SDK?](#what-is-amazon-ads-api-mcp-sdk)\n- [Quick start](#quick-start)\n- [Installation](#installation)\n- [Configuration](#configuration) (auth, packages, profiles, regions)\n- [Downloading reports and exports](#downloading-reports-and-exports)\n- [Example MCP client (Claude Desktop)](#example-mcp-client-connect-claude-desktop)\n- [Context limits](#context-limits-and-active-mcp-server-tools)\n- [Code mode](#code-mode)\n- [Tool audit](#tool-audit)\n- [Background tasks](#background-tasks)\n- [Frequently asked](#frequently-asked)\n- [Troubleshooting](#troubleshooting)\n- [Documentation map](#documentation-map)\n\n## Where this fits in Kuudo\n\nThis repository is the **tool layer** of a broader Amazon agent platform. Kuudo gives Amazon agents tools, knowledge, workflows, and a place to work. This repository is the tools.\n\n| Layer | What it provides | Where |\n| --- | --- | --- |\n| **MCP servers** | The tools an agent calls: Amazon Ads here, plus Selling Partner and Vendor Central | this repo, and the [MCP tool reference](https://www.kuudo.com/docs/mcp-reference/tools/) |\n| **Skills** | The higher-level workflows: reusable, versioned, multi-step Amazon procedures that compose these tools and wait for a human before anything goes live | [Amazon Agent Skills](https://www.kuudo.com/features/skills/) |\n| **Agent Atlas** | The knowledge: indexed Amazon operating knowledge across Ads, Seller Central, and Vendor Central, answered with citations | [Amazon Agent Atlas](https://www.kuudo.com/features/agent-atlas/) |\n| **Agent Flow** | The place to work: your Amazon data landed in your own cloud, so an agent answers with real figures instead of guessing | [Amazon Agent Flow](https://www.kuudo.com/features/amazon-agent-flow/) |\n\nRead this repository as one layer of that platform rather than as the whole of it. The operations below are what an agent can call; Skills are what it runs.\n\n## What Are MCP Tools?\nThink of MCP (Model Context Protocol) as a translator between an AI model and outside systems (like Amazon Ads). Each MCP tool is like a remote control button that tells the AI how to interact with Amazon Ads. Without MCP tools, the AI would have no idea how to “talk” to Amazon Ads.\n\nWith MCP tools:\n* The AI knows the exact endpoints to call.\n* The AI can request campaign reports, budgets, or targeting data safely.\n* Everything is structured, so the AI doesn’t break things by making random guesses.\n\n👉 In short: MCP tools = a safe, well-labeled toolkit that lets AI work with the [Amazon Ads API](https://advertising.amazon.com/API/docs/en-us).\n\n\n\n## 🚀 What is Amazon Ads API MCP SDK?\n\nThe Amazon Ads API MCP SDK is an open-source implementation that provides a robust foundation for creating AI-powered advertising tools, chatbots, and automated services.\n\n### ✨ Key Features\n\n- **🔌 MCP Integration**: Full Model Context Protocol compliance for AI application integration\n- **🌍 Multi-Region Support**: NA, EU, and FE region endpoints with automatic routing\n- **📊 Comprehensive API Coverage**: Campaigns, profiles, reporting, DSP, AMC workflows, and more\n- **📝 Type Safety**: Full Pydantic model support with comprehensive type hints\n- **🧪 Production Ready**: Includes testing, validation, and error handling\n\n## 🎯 Use Cases\n\n### Claude Desktop Integration\n- **Campaign Management**: Ask Claude to create, update, or analyze campaigns\n- **Performance Insights**: Get AI-powered analysis of your advertising performance\n- **Budget Optimization**: Let Claude suggest budget adjustments based on performance\n- **Creative Testing**: Get recommendations for ad creative improvements\n- **Reporting**: Generate custom reports and insights on demand\n\n### AI Applications\n- **Marketing Chatbots**: Build conversational AI that can manage Amazon Ads campaigns\n- **Automated Reporting**: AI-powered insights and performance analysis\n- **Smart Budget Management**: Intelligent budget optimization using AI\n- **Creative Optimization**: AI-driven ad creative testing and optimization\n\n### Enterprise Services\n- **Marketing Automation Platforms**: Integrate Amazon Ads into existing marketing tools\n- **Agency Management Systems**: Multi-client, multi-account advertising management\n- **E-commerce Integrations**: Connect Amazon Ads with e-commerce platforms\n- **Analytics Dashboards**: Real-time advertising performance monitoring\n\n### Developer Tools\n- **API Wrappers**: Create custom SDKs for specific use cases\n- **Testing Frameworks**: Automated testing for Amazon Ads integrations\n- **Development Tools**: Local development and debugging utilities\n\n\n## Why this Ads MCP is different\n\nOther Amazon Ads MCPs do happy path. They fall apart the moment an agent makes a reasonable mistake and Amazon's Ads API has more reasonable mistakes than almost any other large API surface. The v1 reporting catalog uses a different vocabulary than v3. The docs are scattered across migration generations. The same field can have three plausible names depending on which tutorial an agent or human was trained on.\n\n### You've felt this before\n\nYou know the symptoms even if you've never thought of them as a single problem:\n\n- \"Compacting conversation…\" right when you were getting somewhere\n- \"Usage limit reached\" mid-task, with nothing to show for the tokens\n- \"Context window exceeded\" start over, re-explain everything\n- The agent quietly forgetting the fix it figured out three turns ago\n- Five tool calls, four of them retries, none of them progress\n- Re-pasting the same error into the chat for the third time\n\nThat's not the model being dumb. That's the **API surface burning context** vague errors trigger thrashing, thrashing fills the window, the window fills and everything useful gets evicted. Anywhere from 4 to 12 round-trips for an agent: wrong field names, wrong body shape, wrong filter operator, wrong date placement, wrong advertiser-account format.\n\n### Here, the wrong moves return the right move\n\nThis server is built around a different premise: **the error surface is the documentation surface.** Every failure is a teaching opportunity, and the server is engineered to make the next attempt smarter than the last one, for the same agent or the next one, without retraining, prompt updates, or memory tricks.\n\n- The server tells you what went wrong, and *why* you got it wrong\n- Every failure converts to a corrective action — the model just reads the hint and moves on\n- A five-tool-call debugging arc collapses into a two-call corrected one\n- The same mistake a week later doesn't re-burn context, because the validator, alias table, and deprecated-shape table are still right there\n- No out-of-context tribal knowledge to forget\n\nEvery failure is cheaper than the last. Every agent learns from the same authoritative source. Every retry has direction. That compounding is what a smart agent solution looks like in practice.\n\n### Everything is an edge case, you can help\n\nWe won't catch every edge case. Amazon's API surface is enormous, the migration history is messy, and real-world failures get more creative than anything anyone can anticipate. What we're committing to is the **strategy** that errors should teach, that documentation should live in the surfaces agents actually touch, that every wrong move should make the next move easier.\n\nIf you hit a failure where the envelope didn't help; vague hint, wrong suggestion, no hint at all, that's exactly the feedback we want. **Open an issue, paste the envelope, tell us what you expected.** The strategy is only as good as the cases it covers.\n\n## Quick start\n\n**Prerequisites:** [Docker](https://docs.docker.com/get-started/) (recommended), Python 3.10+ if you run from source, and [Amazon Ads API access](https://advertising.amazon.com/API/docs/en-us/guides/get-started/overview) (your own developer app or a partner such as Openbridge).\n\n1. `git clone https://github.com/KuudoAI/amazon-ads-mcp.git && cd amazon-ads-mcp`\n2. `cp .env.example .env` and add credentials (see [Configuration](#configuration)).\n3. `docker compose up -d`\n4. Connect your MCP client to `http://localhost:/mcp/` — `.env.example` sets **`PORT=9080`** (override in `.env` if needed).\n\nOAuth steps, client JSON samples, and a full variable reference: [Configuration](#configuration) below, [**INSTALL.md**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/INSTALL.md), and [**AGENTS.md**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/AGENTS.md).\n\n## 📚 What Is Included In the Amazon Ads MCP?\nThe MCP server mirrors broad coverage of the Amazon Ads API surface. Each enabled package maps to a set of API operations. That includes [Campaign Management (Amazon Ads API v1)](https://advertising.amazon.com/API/docs/en-us/guides/campaign-management/overview), [Exports](https://advertising.amazon.com/API/docs/en-us/guides/exports/overview), [Amazon Marketing Cloud](https://advertising.amazon.com/API/docs/en-us/guides/amazon-marketing-cloud/overview), and many more.\n\nHere is a representative list of the various Amazon API services in the MCP:\n\n- Accounts\n- Audiences\n- Reporting\n- Brand metrics\n- Sponsored Products\n- Sponsored Brands\n- Sponsored Display\n- Amazon DSP\n- Amazon Attribution\n- Recommendations & insights\n- Creatives\n- Change history\n- Data provider\n- Products\n- Unified pre-moderation\n- Moderation\n- Amazon Marketing Stream\n- Locations\n- Exports\n- Media Planning\n- Amazon Ads API v1\n\n### 🧪 Amazon Ads API v1\n\nThe Amazon Ads API v1 represents a reimagined approach to the Amazon Ads API, built from the ground up to provide a seamless experience across all Amazon advertising products through a common model. One major benefit of this common model is improved compatibility with code generation tools such as client library generators.\n\n> **⚠️ Beta Notice**: These APIs are currently in beta at Amazon. Features and endpoints may change. Use in production with caution.\n\n| Package Name | Description | Prefix |\n|-------------|-------------|--------|\n| `ads-api-v1-all` | Ads API v1 merged ALL surface | `allv1_` |\n| `ads-api-v1-sp` | Sponsored Products v1 | `spv1_` |\n| `ads-api-v1-sb` | Sponsored Brands v1 | `sbv1_` |\n| `ads-api-v1-dsp` | Amazon DSP v1 | `dspv1_` |\n| `ads-api-v1-sd` | Sponsored Display v1 | `sdv1_` |\n| `ads-api-v1-st` | Sponsored Television v1 | `stv1_` |\n| `ads-api-v1-beta` | Ads API v1 merged BETA surface | `beta_` |\n\nTo activate Ads API v1 packages, add them to your `AMAZON_AD_API_PACKAGES` environment variable:\n\n```bash\n# Example: Enable the merged Ads API v1 ALL surface\nAMAZON_AD_API_PACKAGES=\"profiles,ads-api-v1-all\"\n```\n\nFor more information, see Amazon's [Campaign Management Overview](https://advertising.amazon.com/API/docs/en-us/guides/campaign-management/overview).\n\n## More than an MCP server\n\nEverything above is the tool layer: the Amazon Ads operations an agent can call. Kuudo adds three layers around it, and a comparison that only counts operations misses them.\n\n| Capability | An MCP server alone | Kuudo |\n| --- | --- | --- |\n| Expose Amazon Ads API operations | yes | yes |\n| Higher-level, multi-step workflows | no | yes, as [Skills](https://www.kuudo.com/features/skills/) |\n| Reusable and versioned workflows | no | yes, each Skill carries a version |\n| Human approval before anything goes live | no | yes, activation is approval-gated |\n| Fork a workflow and change it for your account | no | yes, Skills are a plain open format |\n| Indexed Amazon knowledge, answered with citations | no | yes, through [Agent Atlas](https://www.kuudo.com/features/agent-atlas/) |\n| Your own account data, landed in your own cloud | no | yes, through [Agent Flow](https://www.kuudo.com/features/amazon-agent-flow/) |\n| Workflows that span several MCP servers | no | yes, Ads with Selling Partner and Vendor Central |\n| Run from any MCP client | varies | yes, Claude, ChatGPT, Cursor, and others |\n\n**Kuudo provides higher-level workflows above the MCP tool layer through Skills.** A Skill composes these operations into a repeatable Amazon procedure, grounds it with Atlas, and waits for a human before it changes anything.\n\n## Installation\n\nWe recommend running Amazon Ads API MCP with 🐳 [Docker](https://docs.docker.com/get-started/). Build the image from this repository (there is no supported `docker pull` for a third-party registry image here—use the `Dockerfile` and `docker-compose.yaml` in the repo).\n\nFor bespoke installs such as pip, source checkouts, custom virtualenvs,\nnon-Docker deployments, or Code Mode extra handling, use the\n[**Installation Guide**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/INSTALL.md). The README keeps the Docker quick path\nshort; `INSTALL.md` has the full install matrix and verification steps.\n\n```bash\ngit clone https://github.com/KuudoAI/amazon-ads-mcp.git\ncd amazon-ads-mcp\n```\n\nCopy the environment template:\n\n```bash\ncp .env.example .env\n```\n\nEdit `.env` with your settings (including `PORT` if you change the default).\n\nStart the server with Docker Compose (builds and tags `amazon-ads-mcp:latest` per `docker-compose.yaml`):\n\n```bash\ndocker compose up -d\n```\n\nThe server listens on the host port mapped from `PORT` in `.env` (`.env.example` uses `9080`).\n\nCheck logs:\n\n```bash\ndocker compose logs -f\n```\n\nStop the server:\n\n```bash\ndocker compose down\n```\n\nFor verification, upgrading, developer setup, and non-Docker installation\npaths, see the [**Installation Guide**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/INSTALL.md).\n\n## Configuration\n\n**Operators / self-hosters:** follow the steps below and your `.env` file. **Chat users** mostly interact via prompts once the server is connected; skim [Advertiser Profiles & Regions](#advertiser-profiles--regions) for profile and region behavior.\n\nAmazon Ads requires that all calls to the API are authorized. If you are not sure what this means, you should read the Amazon docs:\n\n* Amazon Ads API onboarding overview: https://advertising.amazon.com/API/docs/en-us/guides/onboarding/overview\n* Getting started with the Amazon Ads API: https://advertising.amazon.com/API/docs/en-us/guides/get-started/overview\n\nThere are three paths for connecting to the API;\n1. Bring Your Own App (BYOA)\n2. Use OpenBridge-managed authorizations\n3. Use Kuudo-managed authorizations\n\n## Bring Your Own Amazon Ads API App\nIf you have your own Amazon Ads API app, or want to create one, the process is detailed below.\n\n### 1. Register Your Application with Amazon\n\n1. Go to the [Amazon Developer Console](https://developer.amazon.com/)\n2. Create or select your Login with Amazon application\n3. Note your `Client ID` and `Client Secret`\n4. Set your callback URL to \"Allowed Return URLs\". This must match the host and port of this MCP server’s HTTP listener (see `PORT` in `.env`; `.env.example` uses **9080**):\n - For production: `https://your-server.com/auth/callback`\n - For local development: `http://localhost:/auth/callback` (e.g. `http://localhost:9080/auth/callback`)\n\nOnce you have your app secured and approved by Amazon, you will need the client ID and secret:\n```bash\n# Amazon Ads API Credentials (required)\nAMAZON_AD_API_CLIENT_ID=\"your-client-id\"\nAMAZON_AD_API_CLIENT_SECRET=\"your-client-secret\"\n```\nMake sure these are in your `.env` file. Also, make sure you set your authorization method to `direct` in the same `.env`:\n\n```bash\nAUTH_METHOD=direct\n```\n\n### Complete OAuth Flow\nTo authorize your connection to Amazon, you need to complete an OAuth workflow as an end user. First, you need to set your region. Authorization occurs at the region level and not setting your region may cause a failure. The server will default to the `na` region. You can manually set the region with tool `set_active_region`.\n\n* Tool: `set_active_region`\n* Parameters: `na` | `eu` | `fe`\n\nExample prompt: *\"Set my current region to `eu`\"*\n\n### Step 1: Start OAuth \n\nTo connect to Amazon Ads API, you use an MCP tool to start your OAuth flow\n* Tool: `start_oauth_flow`\n* Example prompt: *\"Start my OAuth flow\"*\n\n $\"Step$ \n\n### Step 2: Redirect to Amazon Ads\n\nIn this example, you are prompted to click the link that will open a browser window and request approval at Amazon.\n\n $\"Step$ \n\n### Step 3: Approve Request\n\nIn the browser window, Amazon will prompt that you approve the request to connect.\n\n $\"Step$ \n\n### Step 4: Success\n\nIf all goes well, you will see a success response. You can close the browser window and go back to your client. If you see something else, attempt the process again and confirm all your configuration elements are correct\n\n $\"Step$ \n\n\n### Step 5: Confirmation\n\nTo confirm that your MCP server is connected to the Amazon Ads API, check your OAuth status\n\n* Tool: `check_oauth_status`\n* Example prompt: *\"Check my OAuth status\"*\n\n $\"Step$ \n\nYou are ready to start interacting with the Amazon Ads API system!\n\n### Partner Applications: Token Authentication\nYou can configure your client, like Claude, to use authentication by supplying a valid access token. This is most appropriate for service accounts, long-lived API keys, CI/CD, applications where authentication is managed separately, or other non-interactive authentication methods.\n\n#### Openbridge Partner App\nAs an Ads API Partner application provider, Openbridge offers a ready-to-go gateway to the Amazon Ads API. You log into your Openbridge account, provision a token, then set your token in your client config (see below).\n\nFirst, set Openbridge as the auth method:\n\n```bash\nAUTH_METHOD=openbridge\n```\n\nThat is it for the server config. To access the server, you need configure the client, like Claude Desktop, to pass the token directly. (see [Example MCP Client: Connect Claude Desktop](#example-mcp-client-connect-claude-desktop))\n\n#### Authorized Amazon Accounts\n\nYour Amazon authorizations reside in Openbridge. Your first step in your client is to request your current identities: `\"List my remote identities\"`. Next, you would tell the MCP server to use one of these identities: `\"Set my remote identity to <>\"`. You can then ask the MCP to `List all of my Amazon Ad profiles` linked to that account. If you do not see an advertiser listed, set a different identity.\n\n#### Kuudo Partner App\n\nKuudo can vend short-lived Amazon Ads credentials from organization-managed\nAmazon connections. Configure the server with a Kuudo M2M API key:\n\n```bash\nAUTH_METHOD=kuudo\nKUUDO_API_BASE_URL=https://amazon-spapi-dev.kuudo.ai/\nKUUDO_API_KEY=sk_xxxxxx\nKUUDO_PROVIDER=amazon_ads\n```\n\nThe API key requires the `amazon-connections:read` and `amazon-tokens:vend`\nscopes. Set `KUUDO_REMOTE_IDENTITY_ID` to select a default connection at\nstartup, or use the identity tools to list and select a connection before\ncalling Amazon Ads operations.\n\nPublic HTTP deployments must also configure `MCP_INBOUND_TOKEN` or trusted\nproxy HMAC authorization for MCP callers.\n\n\n### Set Your Amazon Ads MCP Packages \nTo activate specific packages, set a comma-separated package list. For a streamlined default with unified Ads API v1 coverage:\n - `AMAZON_AD_API_PACKAGES=\"profiles,accounts-ads-accounts,reporting-version-3,amc-workflow,ads-api-v1-all\"`\n\nHere is the list of tool packages available in the server:\n\n- `profiles`\n- `test-account`\n- `forecasts`\n- `brand-stores-management`\n- `campaign-manage`\n- `accounts-manager-accounts`\n- `accounts-ads-accounts`\n- `accounts-portfolios`\n- `accounts-billing`\n- `accounts-account-budgets`\n- `audiences-discovery`\n- `reporting-version-3`\n- `brand-benchmarks`\n- `brand-metrics`\n- `stores-analytics`\n- `sponsored-products`\n- `sp-suggested-keywords`\n- `sponsored-brands-v4`\n- `sponsored-brands-v3`\n- `sponsored-display`\n- `dsp-measurement`\n- `dsp-advertisers`\n- `dsp-audiences`\n- `dsp-conversions`\n- `dsp-target-kpi-recommendations`\n- `amazon-attribution`\n- `audience-insights`\n- `partner-opportunities`\n- `tactical-recommendations`\n- `persona-builder`\n- `creative-assets`\n- `change-history`\n- `data-provider-data`\n- `products-metadata`\n- `products-eligibility`\n- `unified-pre-moderation-results`\n- `moderation-results`\n- `amazon-marketing-stream`\n- `locations`\n- `exports-snapshots`\n- `marketing-mix-modeling`\n- `reach-forecasting`\n- `amc-administration`\n- `amc-workflow`\n- `amc-rule-audience`\n- `amc-ad-audience`\n- `ads-api-v1-sp` *(Beta)*\n- `ads-api-v1-sb` *(Beta)*\n- `ads-api-v1-dsp` *(Beta)*\n- `ads-api-v1-sd` *(Beta)*\n- `ads-api-v1-st` *(Beta)*\n- `ads-api-v1-all`\n- `ads-api-v1-beta` *(Beta)*\n\nYou will note that some are broken up into smaller groupings. For example, Amazon Marketing Cloud has bundles; `amc-ad-audience`, `amc-administration`, `amc-rule-audience`, and `amc-workflow`. This is done to create efficiencies and optimizations that reduce context limits in many AI clients. \n\n## Understanding Amazon Ads MCP Tools \n\nAmazon Ads MCP tools have prefixes (like `cm_` for Campaign Management or `amc_` for Amazon Marketing Cloud) to help organize the specific Ads API operation.\n\nExample prefixes:\n- `cm_` → campaign/advertising APIs\n- `amc_` → AMC-related APIs\n- `dsp_` → DSP APIs\n- `sd_` → Sponsored Display\n- `ams_` → Amazon Marketing Stream\n- `spv1_` → Sponsored Products v1 *(Beta)*\n- `sbv1_` → Sponsored Brands v1 *(Beta)*\n- `dspv1_` → Amazon DSP v1 *(Beta)*\n- `sdv1_` → Sponsored Display v1 *(Beta)*\n- `stv1_` → Sponsored Television v1 *(Beta)* \n\nThis will translate into collections of tools that align with the API operations that are available:\n\n**Campaign Management (`cm_`)**\n- `cm_QueryCampaign` — Query campaigns \n- `cm_CreateCampaign` — Create campaign \n- `cm_UpdateCampaign` — Update campaign \n- `cm_DeleteCampaign` — Delete campaign \n- `cm_QueryAdGroup` — Query ad groups \n\n**Sponsored Products (`sp_`)**\n- `sp_listProductAds` — List product ads \n- `sp_createKeywords` — Create keywords \n- `sp_updateBids` — Update keyword bids \n- `sp_getNegativeKeywords` — Get negative keywords \n\n**AMC Workflows (`amc_`)**\n- `amc_listWorkflows` — List AMC workflows \n- `amc_executeWorkflow` — Run workflow \n- `amc_getWorkflowStatus` — Check workflow status \n\nUsers would see tools like:\n\n- **\"List my Amazon Ads campaigns\"** \n → Claude uses: `cm_QueryCampaign`\n\n- **\"Create an AMC workflow\"** \n → Claude uses tools such as `amc_executeWorkflow` (after discovery; exact names depend on your enabled packages)\n\n- **\"Export my sponsored products ads data\"**\n → Claude uses: `export_createAdExport`\n\n\n\n## 📥 Downloading Reports & Exports\n\nWhen tools fetch a report or export from Amazon, they **download the file onto the machine running the MCP server** (local disk on the server host—Docker volume, VM, or bare metal—not your laptop unless the server runs there). The default base directory is `./data` (override with `AMAZON_ADS_DOWNLOAD_DIR`). Files are organized under `data/profiles/{profile_id}/…`.\n\nMCP tools list those server-side files and build URLs; clients then pull the bytes over **HTTP** (`GET /downloads`, `GET /downloads/{relative-path}`). Nothing streams straight from Amazon into the MCP client without hitting server storage first in these flows.\n\n**Use HTTP transport** for downloads: run with `--transport http` (or Docker on port 9080). **stdio** mode does not register `/downloads`, and the `get_download_url` tool returns an error without HTTP.\n\n### Reports vs exports\n\n- **V3 async reports:** Use the OpenAPI tools `createAsyncReport` (POST) to create, `getAsyncReport` (GET) to poll status, then `download_export` to save the completed file **on the server** into profile storage.\n- **Exports (e.g. ad exports):** Create the job with the OpenAPI tools (e.g. `export_CampaignExport`), poll with `export_GetExport`, then call `download_export` with the completion URL so the file is **fetched from Amazon and written on the server** (same storage layout as reports).\n\n### Workflow\n\n1. **Set active profile** (`set_active_profile`) so storage and HTTP access are scoped.\n2. **Create the job:** Use the appropriate OpenAPI tool to create a report or export.\n3. **Poll for completion:** Use the corresponding GET tool to check status until `COMPLETED`.\n4. **Download:** Call `download_export` with the download URL from the completed response.\n5. **Discover or link:** MCP tools `list_downloads` and `get_download_url`.\n6. **Fetch bytes:** HTTP `GET` on the URL from step 5 (browser, curl, etc.).\n\nSubpaths under the profile folder depend on report/export type (for example `reports/…`, `exports/…`). URLs look like `http://localhost:9080/downloads/{relative-path}` where `{relative-path}` matches `list_downloads` (illustrative example: `reports/async/some-report.json.gz`).\n\n**Direct HTTP (`curl`):** `/downloads` uses the **active profile** in the running server (the one set via MCP in that process). If you see `No active profile`, call `set_active_profile` through MCP against the same server first. When no auth manager is configured, a numeric `?profile_id=` query parameter may be accepted instead—see your deployment settings.\n\n### Example Prompts\n\n| Task | Example Prompt |\n|------|----------------|\n| Download a report | *\"Generate a Sponsored Products report for January 2026\"* |\n| List available files | *\"Show me my downloaded files\"* |\n| Get download link | *\"Get the download URL for the report we just created\"* |\n| Filter by type | *\"List my downloaded campaign exports\"* |\n\n### HTTP Download API\n\n```bash\n# List available downloads (paths and URLs are profile-scoped)\ncurl http://localhost:9080/downloads\n\n# Optional filter\ncurl 'http://localhost:9080/downloads?type=reports'\n\n# Download one file (example path — use paths from list_downloads)\ncurl -O 'http://localhost:9080/downloads/reports/async/example-report.json.gz'\n\n# If AMAZON_ADS_DOWNLOAD_AUTH_TOKEN is set\ncurl -H \"Authorization: Bearer your-token\" -O 'http://localhost:9080/downloads/exports/campaigns/example-export.json'\n```\n\n### Profile Isolation\n\n- Files live under `data/profiles/{profile_id}/`.\n- Download tools and `/downloads` only see the **active** profile’s files.\n- Set the profile first: *\"Set my active profile to 123456789\"*.\n- Legacy files outside `profiles/` are not listed when using profile-scoped mode; move them under `data/profiles/{profile_id}/` for access.\n\n## Advertiser Profiles & Regions\n\n### Setting Your Advertiser Profile \nPer Amazon: *Profiles play a crucial role in the Amazon Ads API by determining the management scope for a given call. A profile ID is a required credential to access an advertiser's data and services in a specific marketplace.*\n\nYou may not know what profile(s) authorization grants you access to. You can list all advertising profiles accessible by your authorization:\n\n* Tool: `ac_listProfiles`\n* Example prompt: *\"List my advertiser profile ids\"*\n\n**Warning:** Large accounts can return very large profile lists that may exceed client context limits. Prefer these bounded tools for discovery:\n\n* Tool: `summarize_profiles` — *\"Summarize my advertiser profiles\"*\n* Tool: `search_profiles` — *\"Find profiles with Acme in the name in US\"*\n* Tool: `page_profiles` — *\"Show the first 20 UK profiles\"*\n* Tool: `refresh_profiles_cache` — *\"Refresh my profile list cache\"*\n\nResponse includes profile details:\n- profileId, countryCode, currencyCode\n- dailyBudget, timezone\n- accountInfo (type: seller/vendor/agency)\n\nLet's assume your list included profile ID `1043817530956285`. You can check for more details by getting profile details to confirm this is the one you want to use.\n\n* Tool: `ac_getProfile` \n* Example prompt: *\"Get the details for my profile_id: `1043817530956285`\"*\n\nAssuming this is the profile you want to use, you need to **set** the profile Amazon requires for API calls:\n* Tool: `set_active_profile`\n* Example prompt: *\"Set my active profile id to `1043817530956285`\"*\n\nWhen you set the profile, it determines:\n- Which account's data you access\n- Currency and timezone for reports\n- Available campaigns/ads/keywords\n\nThe profile ID will be set in the background for the duration of your session. Repeat the process if you want to switch to a new profile.\n\nMost calls to the Amazon Ads API require a Region. Each [advertiser profile ID](https://advertising.amazon.com/API/docs/en-us/guides/account-management/authorization/profiles) is associated with an advertising account in a specific region/marketplace. \n\nThe region is part of an advertiser profile. When you set an advertiser profile with `set_active_profile`, it will set the region that is associated with the profile automatically.\n\n* Tool: `set_active_profile`\n\nExample prompt: *\"Set my active advertiser profile to `111111111111`\"* \n\nSince profile ID `111111111111` is based in `na`, the region will be set based on the profile region.\n\n### Set Active Region\nThe Amazon Ads MCP server includes tools for managing API regions as defaults and dynamically, allowing you to switch between North America (`na`), Europe (`eu`), and Far East (`fe`) regions without restarting the server.\n\n| Region Code | Name | API Endpoint \n|------------|------|--------------|\n| `na` | North America | https://advertising-api.amazon.com\n| `eu` | Europe | https://advertising-api-eu.amazon.com\n| `fe` | Far East | https://advertising-api-fe.amazon.com \n\nWhen you set a region, the system automatically:\n\n1. **Updates API endpoints** - Routes API calls to the correct regional endpoint\n2. **Updates OAuth endpoints** - Uses the correct token refresh endpoint for the region\n3. **Clears cached tokens** - Ensures fresh authentication for the new region\n4. **Preserves other settings** - Keeps profile ID and identity settings intact\n\n\n**IMPORTANT: Avoid Region Mismatch**: *If you attempt to set a region that is not associated with your advertiser profile, the Ads API will reject your requests. For example, if a profile ID is attached to `na` and you manually set the region to `eu`, you have created a mismatch which will cause API request failures.*\n\n### Get Active Region\nIf you are not sure what region is set, you can check for the region\n* Tool: `get_active_region`\n* Returns: Current region, endpoints, and configuration source\n\nExample prompt: *\"What is my current active region?\"*\n\n## Example MCP Client: Connect Claude Desktop\n\nNavigate to Connector Settings\n\nOpen Claude in your browser and navigate to the settings page. You can access this by clicking on your profile icon and selecting “Settings” from the dropdown menu. Once in settings, locate and click on the “Connectors” section in the sidebar. This will display your currently configured connectors and provide options to add new ones.\n\nEdit your Claude Desktop configuration file:\n\n**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json` \n**Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json` \n**Linux**: `~/.config/Claude/claude_desktop_config.json`\n\nIn this example, we show how to use the bearer token using the Openbridge API key. Add this configuration to your `mcpServers` section:\n\n```json\n{\n \"mcpServers\": {\n \"amazon_ads_mcp\": {\n \"command\": \"npx\",\n \"args\": [\n \"-y\",\n \"mcp-remote@latest\",\n \"http://${HOSTNAME}:${PORT}/mcp/\",\n \"--allow-http\",\n \"--header\",\n \"Authorization:Bearer ${OPENBRIDGE_API_KEY}\",\n \"--header\",\n \"Accept:application/json,text/event-stream\",\n \"--debug\"\n ],\n \"env\": {\n \"HOSTNAME\": \"your_hostname\",\n \"PORT\": \"your_server_port\",\n \"MCP_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_CONNECTION_TIMEOUT\": \"10000\",\n \"MCP_SERVER_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_TOOL_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_WARNING_THRESHOLD\": \"10000\",\n \"OPENBRIDGE_API_KEY\": \"your_openbridge_token_here\"\n }\n }\n }\n}\n```\n**Note**: Replace `hostname`, `port` and `your_openbridge_token_here` with your actual OpenBridge token.\n\n**IMPORTANT**: Cursor and Claude Desktop (Windows) have a bug where spaces inside args aren't escaped when it invokes npx, which ends up mangling these values. You can work around it using: [mcp-remote custom headers documentation](https://github.com/geelen/mcp-remote?tab=readme-ov-file#custom-headers). \n\nThe config would look something like this:\n\n\n```json\n{\n \"mcpServers\": {\n \"amazon_ads_mcp\": {\n \"command\": \"npx\",\n \"args\": [\n \"-y\",\n \"mcp-remote\",\n \"http://${HOSTNAME}:${PORT}/mcp/\",\n \"--allow-http\",\n \"--header\",\n \"Authorization:${AUTH_HEADER}\",\n \"--header\",\n \"Accept: application/json, text/event-stream\"\n ],\n \"env\": {\n \"HOSTNAME\": \"your_hostname\",\n \"PORT\": \"your_server_port\",\n \"MCP_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_CONNECTION_TIMEOUT\": \"10000\",\n \"MCP_SERVER_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_TOOL_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_WARNING_THRESHOLD\": \"10000\",\n \"AUTH_HEADER\": \"Bearer \"\n }\n }\n }\n}\n```\n\nHere is another example, which can be used if you are using OAuth since the `OPENBRIDGE_API_KEY` is not needed:\n\n```json\n{\n \"mcpServers\": {\n \"amazon_ads_mcp\": {\n \"command\": \"npx\",\n \"args\": [\n \"-y\",\n \"mcp-remote@latest\",\n \"http://localhost:9080/mcp/\",\n \"--allow-http\"\n ],\n \"env\": {\n \"MCP_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_CONNECTION_TIMEOUT\": \"10000\",\n \"MCP_SERVER_REQUEST_TIMEOUT\": \"60000\",\n \"MCP_TOOL_TIMEOUT\": \"120000\",\n \"MCP_REQUEST_WARNING_THRESHOLD\": \"10000\"\n }\n }\n }\n}\n```\n\n*Note: For various Claude configurations similar to what was shown above, see the [MCP Remote docs](https://github.com/geelen/mcp-remote) for the latest settings/options.*\n\n### Restart Claude Desktop\n\nAfter saving the configuration file, restart Claude Desktop to load the new MCP server.\n\n\n\n## ⚠️ Context Limits and Active MCP Server Tools\nMCP tool registration and use can impact your AI systems usage limits. Usage limits control how much you can interact with an AI system, like Claude, over a specific time period. As Anthropic states, think of the amount of information/data used as drawing down on a \"conversation budget\". That budget determines how many messages you can send to your AI client, or how long you can work, before needing to wait for your limit to reset.\n\nMCP Server tools contribute metadata like titles, descriptions, hints, and schemas to the model's context. This metadata is loaded into the LLM’s context window, which acts as its short-term working memory.\n\nEach client, like Claude, has a fixed-size context window. This defines the maximum amount of information it can process in a single interaction—including user prompts, system instructions, tool metadata, and any prior messages. \n\nThe more tools you activate, the more of that limited space gets consumed up front. When you activate many tools, their combined schema and config payloads can significantly use up this context and you may quickly hit the context ceiling. This is when you’ll start seeing errors or warnings about exceeding the chat length limit.\n\n**The Amazon Ads MCP provides coverage across the entire API. As a result, there can be 100s of tools!**\n\n* More tools = less room for user interaction: Activating unnecessary tools reduces available space for your actual prompt or data.\n* Start small: Activate only what you need for the current task. You can always add more later.\n\nIf you're encountering unexpected length issues, review which tools are active. Trimming unused ones can help minimize context use.\n\n## Code Mode\n\nCode Mode is a feature that dramatically reduces tool token consumption. Instead of loading every tool schema into the LLM's context upfront, Code Mode replaces the full catalog with four lightweight meta-tools. The LLM discovers tools on demand and executes them via sandboxed Python.\n\n### Token Impact\n\n| Mode | Tools in Context | Tokens Used | Context Window |\n|------|-----------------|-------------|----------------|\n| Standard | 55 | ~32,000 | 16.1% |\n| Code Mode | 4 | ~470 | 0.2% |\n| **Reduction** | | | **98.6%** |\n\n*Illustrative comparison for a representative default-sized catalog; actual tool counts and token use depend on `AMAZON_AD_API_PACKAGES`, server version, and tokenizer.*\n\n### How It Works\n\nCode Mode uses a 4-stage discovery pattern:\n\n```\nStage 1: tags → Browse categories: \"campaign-management (20), accounts (14), ...\"\nStage 2: search → Find tools: search(\"campaigns\") → cm_listCampaigns, cm_createCampaign, ...\nStage 3: get_schema → Get parameters: get_schema(\"cm_listCampaigns\") → full input schema\nStage 4: execute → Run code: sandboxed Python with await call_tool(\"cm_listCampaigns\", {...})\n```\n\nAll 200+ tools remain fully accessible. The LLM simply discovers and calls them on demand rather than loading every schema upfront.\n\n### Activating Code Mode\n\nCode Mode defaults to **`CODE_MODE=true`**. To use the full tool catalog in context instead, set:\n\n```bash\nCODE_MODE=false\n```\n\n**Docker Compose** — add to your `environment` section:\n```yaml\nenvironment:\n - CODE_MODE=false\n```\n\n**Docker Run** (map the host port to the container `PORT`, e.g. 9080):\n```bash\ndocker run -d --env-file .env -e CODE_MODE=false -e PORT=9080 -p 9080:9080 amazon-ads-mcp:latest\n```\n\n**Local Development** (entry point is `python -m amazon_ads_mcp.server`, which runs the MCP server module):\n```bash\nCODE_MODE=false uv run python -m amazon_ads_mcp.server --transport http --port 9080\n```\n\n### Configuration Options\n\n| Variable | Default | Description |\n|----------|---------|-------------|\n| `CODE_MODE` | `true` | Enable code mode |\n| `CODE_MODE_INCLUDE_TAGS` | `true` | Include tag browsing in discovery (set `false` for small catalogs) |\n| `CODE_MODE_MAX_DURATION_SECS` | `30` | Maximum execution time per sandbox run |\n| `CODE_MODE_MAX_MEMORY` | `50000000` | Memory limit per sandbox run (50 MB) |\n\n### Using Code Mode\n\nOnce enabled, your MCP client (Claude Desktop, Claude Code, etc.) will see only four tools instead of the full catalog.\n\n**Browse available categories:**\n> *\"What tool categories are available?\"*\n\nThe LLM calls `tags` and sees categories like `campaign-management`, `sponsored-products`, `programmatic-dsp`, `accounts`, `reporting`, etc.\n\n**Find specific tools:**\n> *\"Find tools for managing campaigns\"*\n\nThe LLM calls `search(\"campaigns\")` and gets matching tool names with brief descriptions.\n\n**Get tool details:**\n> *\"Show me the parameters for creating a campaign\"*\n\nThe LLM calls `get_schema(\"cm_CreateCampaign\")` and gets the full input schema.\n\n**Execute operations:**\n> *\"List my enabled campaigns\"*\n\nThe LLM calls `execute` with Python code:\n```python\nresult = await call_tool(\"cm_QueryCampaign\", {\n \"body\": {\"stateFilter\": {\"include\": [\"ENABLED\"]}}\n})\n```\n\n### Tool Categories\n\nWhen the LLM calls `tags`, it sees human-readable categories mapped from the API prefixes:\n\n| Category | Prefix | Description |\n|----------|--------|-------------|\n| `campaign-management` | `cm` | Create, update, query, delete campaigns, ad groups, ads, targets |\n| `sponsored-products` | `sp` | Sponsored Products targeting, keywords, bids |\n| `sponsored-brands` | `sb` | Sponsored Brands campaigns (v3 + v4) |\n| `sponsored-display` | `sd` | Sponsored Display campaign and targeting tools |\n| `programmatic-dsp` | `dsp` | DSP advertisers, audiences, conversions, measurement |\n| `amazon-marketing-cloud` | `amc` | AMC workflows, audiences, administration |\n| `accounts` | `ac` | Profiles, billing, budgets, portfolios, manager accounts |\n| `reporting` | `rp` | V3 reporting tools |\n| `brand-insights` | `br` | Brand benchmarks and metrics |\n| `stores` | `st` | Stores analytics |\n| `server-management` | — | Profile, region, OAuth, download tools |\n\n### When to Use Code Mode\n\n**Use Code Mode when:**\n- You have many tool packages activated and are hitting context limits\n- You want maximum context space for conversation and data analysis\n- Your MCP client supports tool discovery patterns (Claude Desktop, Claude Code)\n\n**Use Standard Mode when:**\n- You have a small number of packages activated (1-3)\n- You need the fastest possible tool invocation (no discovery step)\n- Your workflow uses the same few tools repeatedly\n\n### Verifying Code Mode\n\nAfter starting the server with `CODE_MODE=true`, verify it is active:\n\n```bash\n# Check server logs for confirmation\ndocker logs 2>&1 | grep \"Code mode\"\n# Expected: \"Code mode active: 4 meta-tools exposed (tags, search, get_schema, execute)\"\n```\n\nIn your MCP client, you should see only four tools: `tags`, `search`, `get_schema`, and `execute`. If you see the full tool catalog, code mode is not active — check your environment variable.\n\n## Tool Audit\n\nUse `tool_audit` to measure token footprint for tool catalogs and code mode behavior.\n\n### Baseline Audit\n\n```bash\n.venv/bin/python -m amazon_ads_mcp.tool_audit \\\n --url http://127.0.0.1:9080/mcp \\\n --limit 20\n```\n\nJSON output:\n\n```bash\n.venv/bin/python -m amazon_ads_mcp.tool_audit \\\n --url http://127.0.0.1:9080/mcp \\\n --format json \\\n --limit 0 \\\n > /tmp/tool_audit.json\n```\n\n### Code Mode Probe Audit\n\nWhen code mode is active, use probe mode to estimate on-demand schema fetch cost:\n\n```bash\n.venv/bin/python -m amazon_ads_mcp.tool_audit \\\n --url http://127.0.0.1:9080/mcp \\\n --format json \\\n --limit 0 \\\n --code-mode-probe \\\n --probe-limit 50 \\\n --probe-queries \"campaign,ad,report,profile,dsp,amc,brand,targeting\" \\\n > /tmp/tool_audit_code_mode.json\n```\n\n### Quick Comparison\n\n```bash\njq '{tool_count,total_tool_tokens,context_window_percent}' /tmp/tool_audit.json\njq '{tool_count,total_tool_tokens,context_window_percent,code_mode_probe}' /tmp/tool_audit_code_mode.json\n```\n\nKey fields:\n- `total_tool_tokens`: Upfront token load from `tools/list`\n- `context_window_percent`: Share of configured context window\n- `code_mode_probe.total_schema_tokens`: Total sampled schema-fetch token cost in code mode\n- `code_mode_probe.avg_schema_tokens`: Average token cost per fetched schema\n\n## Background Tasks\n\nMany Amazon Ads API operations are long-running — report generation, export creation, AMC workflow execution, audience processing, and DSP measurement studies can take seconds to minutes. Background tasks allow these operations to run without blocking the conversation.\n\n### How It Works\n\nBackground tasks are **enabled by default**. When a client requests background execution for any tool:\n\n1. **Start**: The tool returns immediately with a task ID\n2. **Track**: The client polls for progress updates at server-suggested intervals\n3. **Retrieve**: The client fetches the result when the task completes\n\nAll async tools in the server support background execution automatically. The client decides per-call whether to run a tool in the foreground (wait for result) or background (get task ID, poll later).\n\n### Built-in workflow tools\n\nThese tools orchestrate multi-step workflows (download and HTTP access to server-side files):\n\n| Tool | Purpose |\n|------|---------|\n| `download_export` | Downloads a completed export or report file from Amazon S3 to local storage |\n| `list_downloads` | Lists all downloaded files for the active profile |\n| `get_download_url` | Returns an HTTP URL to fetch a downloaded file |\n\nThese tools include progress reporting so clients can track each stage of the workflow.\n\n### Configuration\n\n| Variable | Default | Description |\n|----------|---------|-------------|\n| `ENABLE_TASKS` | `true` | Enable background task execution for all async tools |\n\nTo disable background tasks:\n\n```bash\nENABLE_TASKS=false\n```\n\n**Docker Compose** — add to your `environment` section:\n```yaml\nenvironment:\n - ENABLE_TASKS=false\n```\n\n### Backend\n\nThe server uses an in-memory task backend by default. Tasks are tracked for the lifetime of the server process. If the server restarts, in-progress tasks are lost.\n\nFor persistent task tracking across restarts, configure a Redis backend:\n\n```bash\nFASTMCP_DOCKET_URL=redis://localhost:6379\n```\n\n### Example Prompts\n\n| Task | Example Prompt |\n|------|----------------|\n| Generate a report | *\"Generate a Sponsored Products campaign report for March 2026\"* |\n| Export campaigns | *\"Export all my enabled campaigns\"* |\n| Run AMC workflow | *\"Execute my audience overlap workflow for the last 30 days\"* |\n\n## Frequently asked\n\n### Does Kuudo only expose Amazon Ads API operations?\n\nNo. This server exposes the Amazon Ads operations themselves. Kuudo Skills package them into higher-level workflows: reusable, versioned, multi-step procedures that can combine tools from several MCP servers, retrieve Amazon and account knowledge through Agent Atlas, use Agent Flow for durable data work in your own cloud, and wait for human approval before anything goes live. Skills run from any MCP client, from code, or from a CLI.\n\n### Do I have to use a prebuilt Skill?\n\nNo. An agent can compose these operations directly for one-off work. Skills exist for the procedures you run repeatedly and want to run the same way every time. Pick one, watch it run, then fork it for your account, or write your own against these tools: every Skill is a plain open-format file, not a proprietary one.\n\n### Is this the whole product?\n\nNo. This repository is the Amazon Ads tool layer. Kuudo also runs MCP servers for [Selling Partner](https://www.kuudo.com/features/amazon-selling-partner-mcp/) and [Vendor Central](https://www.kuudo.com/features/amazon-vendor-central-mcp/), the Skills workflow layer, Agent Atlas for knowledge, and Agent Flow for durable data work. See [how the pieces fit](https://www.kuudo.com/features/).\n\n## Troubleshooting\n\n**Server not connecting?**\n- Ensure the Docker container is running: `docker compose ps`\n- Check server logs: `docker compose logs -f`\n- Verify the host port matches **`PORT`** in `.env` (`.env.example` uses **9080**; your mapping must match)\n\n**Authentication errors?**\n- Check your OpenBridge token is valid\n- Ensure the token is properly set in the environment\n- Verify your Amazon Ads API access\n\n**Claude not recognizing the server?**\n- Restart Claude Desktop after configuration changes\n- \"Reload Page\" in Claude Desktop if the MCP is not active\n- Check the JSON syntax is valid\n- Ensure the server name matches exactly\n\n## Documentation map\n\n| Doc | Purpose |\n|-----|---------|\n| [**INSTALL.md**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/INSTALL.md) | Full install, verification, upgrades, developer setup |\n| [**AGENTS.md**](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/AGENTS.md) | Environment variables, Docker, MCP tuning, contribution workflow |\n| This README | Overview, auth walkthrough, downloads, clients, code mode |\n| [**Kuudo MCP tool reference**](https://www.kuudo.com/docs/mcp-reference/tools/) | Every Kuudo MCP server and its tools |\n| [**Amazon Agent Skills**](https://www.kuudo.com/features/skills/) | The workflow layer built on these operations |\n\n## 📄 License\n\nThis project is licensed under the MIT License — see the [LICENSE](https://github.com/KuudoAI/amazon_ads_mcp/blob/main/LICENSE) file for details.\n\n**Security:** report vulnerabilities through [GitHub private security advisories](https://github.com/KuudoAI/amazon-ads-mcp/security/advisories/new) for this repository.\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt",1770079621000,"priceCents","trialDays","iconType","image","upstream",{"_107":108,"_109":110,"_111":112,"_113":114,"_115":116,"_117":118},"requiredSecrets",[],"tags",[97,98,99,100,21,101,102,103,104,105,106],"assets",[],"stats",{"_95":44,"_96":44},"ratingAvg","reviewCount","ai-agent","amazon","amazon-ads","amazon-ads-api","amazon-advertising","amazon-advertising-api","chatgpt","claude","claude-code","cursor","topics",[97,98,99,100,21,101,102,103,104,105,106,123,124,125],"license","MIT","language","Python","registryName","io.github.tspicer/amazon_ads_mcp","homepage","https://www.kuudo.com/","packages",[119],{"_120":121,"_122":21,"_72":73},"registryType","pypi","identifier","mcp","mcp-server","model-context-protocol","ENV",{"_128":129,"_130":131,"_132":133,"_134":-7},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272","SENTRY_ENABLE_LOCAL"]