Overview

Model Context Protocol (MCP) is an open standard that allows AI-powered tools to securely access external tools, APIs, and documentation sources. Connecting your AI-powered tool to the Saner API Docs MCP Server gives it contextual awareness of Saner APIs, so it can help you build integrations without switching between documentation and code.

The Saner API Docs MCP Server gives AI-powered tools direct access to the Saner API specifications and Python SDK documentation. Your AI-powered tool can then explore Saner APIs, answer questions about parameters and responses, and generate integration code grounded in the official documentation, all from inside your editor.

Use This When You Want To

Explore Saner APIs using natural-language prompts
Understand request parameters and response structures without leaving your editor
Generate ready-to-use Python SDK examples
Discover related APIs across organizations, agents, vulnerabilities, compliance data, and reports
Build automation scripts faster, with SDK usage patterns grounded in Saner documentation

What the Server Does and Does Not Do

The Saner API Docs MCP Server does	The Saner API Docs MCP Server does not
Serve the complete OpenAPI specification for every Saner API	Call the Saner API
Serve Python SDK documentation and usage examples	Store or ask for your API key
Give your AI-powered tool everything it needs to write a working integration script	Connect to your Saner account or read your tenant data

What You Can Ask Your AI-Powered Tool

Once connected, your AI-powered tool understands which Saner APIs exist and how they map to SDK resources and workflows. Some examples:

Discover an API

Show me how to get organization details in Saner

Look up parameters

What parameters are required for agent download in Saner?

Generate integration code

Generate Python SDK code to export a vulnerability report in Saner

Explore related APIs

Show APIs related to compliance management in Saner

Read-Only by Design

The Saner API Docs MCP Server is read-only. It serves documentation and nothing else:

It does not call the Saner API
It does not store, request, or hold your API credentials
It cannot connect to your Saner account or read your tenant data

Everything it returns comes from the published Saner API specifications and SDK documentation. Because the server never contacts Saner, anything your AI-powered tool tells you about your environment comes from a script you ran yourself, not from this server.

Compatible AI-Powered Tools

The Saner API Docs MCP Server is built on the open Model Context Protocol, so it works with MCP-compatible AI-powered tools, including:

Claude Desktop
Claude Code
LM Studio

Other AI-powered tools that support MCP can connect to the server using their standard MCP configuration.

Before You Run Generated Code

The server itself is read-only, but the code your AI-powered tool generates with it is not.

🚧
Generated scripts run against your live Saner tenant
Saner APIs that create, update, delete, scan, or remediate make real changes to your environment. Always review generated code before you run it.

Before you run a generated script:

Review it end to end. Confirm which endpoints it calls and what each call changes.
Test in a non-production account first, where one is available.
Start with read-only calls. Verify the results before you run anything that modifies data.
Keep your API key out of the code. Load it from an environment variable or a secrets manager rather than hardcoding it in the script.

Next Steps

See MCP Server Setup to set it up in Claude Code, Cursor, Windsurf, or Claude Desktop
See Python SDK Overview for the SDK the generated code will use