Amazon knowledge retrieval MCP server (Amazon Agent Atlas) by Kuudo

A retrieval layer your agent checks before it answers. It indexes Amazon operating knowledge across Ads, Seller Central, and Vendor Central.

Product page: https://www.kuudo.com/features/agent-atlas/ · Docs: https://www.kuudo.com/docs/amazon-agent-atlas/ · Pricing: https://www.kuudo.com/pricing.md

Connect

Kuudo runs in your own cloud. The Community plan deploys one instance of each Amazon MCP server into your account, and your client connects to that deployment:

config.json

{
  "mcpServers": {
    "amazon-agent-atlas-mcp": {
      "url": "https://<your-host>/mcp",
      "headers": {
        "Authorization": "Bearer <your Kuudo API key>"
      }
    }
  }
}

Replace <your-host> with the hostname of your deployment and the bearer value with your Kuudo API key.

A tool may document a provider credential of its own. That is separate from the bearer above, it belongs to the service the tool calls, and the tool's own entry says so.

What this repository is

This repository holds registry metadata and a catalog-only stub. Live execution runs in your Kuudo deployment. The server source is not published. The stub in src/ answers tools/list with the catalog below, serves the same catalog as one resource (kuudo://catalog/tools.json), offers one prompt (connect) carrying the setup guidance, and returns an error with that guidance on any call, so registries and clients can inspect the surface without any access to Amazon.

Inspect the catalog locally with Docker

The image runs the same catalog-only stub over stdio. It is not the live server.

Terminal

docker build -t amazon-agent-atlas-mcp .
docker run -i --rm amazon-agent-atlas-mcp

Point a client at it with a stdio entry:

config.json

{
  "mcpServers": {
    "amazon-agent-atlas-mcp-catalog": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "amazon-agent-atlas-mcp"]
    }
  }
}

Tools

4 tools, read from a running instance of this server. Four tools front the whole corpus: browse tags, search, read a schema, then run a chained call.

Each tool carries the argument schema the live server publishes, so a client can inspect the full call signature here before it connects to your deployment.

Every tool (4)

Tool	Access	What it does
`tags`	read	List available tool tags. Use to browse available tools by tag before searching.
`search`	read	Search for available tools by query. Returns matching tools ranked by relevance.
`get_schema`	read	Get parameter schemas for specific tools. Use after searching to get the detail needed to call a tool.
`execute`	write	Chain `await call_tool(...)` calls in one Python block and return the final value. Only `call_tool(tool_name: str, params: dict) -> Any` is available as an external function. Supported imports: asyncio, datetime, json, math, os, pathlib, re, sys, typing. Other standard-library and third-party imports, including collections, are unavailable in the sandbox. Use f-strings because str.format is unavailable. Limits: 10 seconds, 50,000,000 bytes of memory, and recursion depth 100.

Set up your client

Reads, writes, approvals

Read tools are safe to call freely. Write tools do work inside your deployment and never change anything on Amazon. Each tool's access is recorded beside it in tools.json.

Kuudo

Website: https://www.kuudo.com/
Docs: https://www.kuudo.com/docs/
Guides: https://www.kuudo.com/guides/
Community, bugs, and questions: https://github.com/KuudoAI/community
Roadmap: https://github.com/orgs/KuudoAI/projects
Machine-readable: https://www.kuudo.com/llms.txt · https://www.kuudo.com/pricing.md

Generated from KuudoAI/marketing. Do not edit by hand; changes are overwritten on the next sync.