Kilometers Documentation

Everything you need to monitor your AI agents effectively. Transparent MCP server monitoring with zero configuration changes required.

Documentation Updated

Our documentation has been updated to reflect the actual CLI functionality. The Kilometers CLI works as a transparent wrapper around MCP servers - no separate configuration files or complex setup required.

Quick Start
Get started with Kilometers in just a few commands
1

Install Kilometers CLI

Download the binary from GitHub releases and add to PATH

TERMINAL
# Download from GitHub releases # Extract and add km to PATH
2

Configure environment

Set your API credentials

TERMINAL
export KILOMETERS_API_KEY="your-api-key"
3

Wrap your MCP server

Use km to monitor any MCP server transparently

TERMINAL
km npx @modelcontextprotocol/server-github
Popular Articles
Most helpful guides for getting started with Kilometers
Setup
5 min read

Setting up MCP server monitoring

Complete guide to transparently monitoring any MCP server

Security
4 min read

Understanding risk detection

How Kilometers analyzes and scores MCP events for security

Configuration
3 min read

Environment variable reference

Complete list of KM_* environment variables for configuration

Filtering
6 min read

Filtering MCP events

Use method whitelisting and payload limits to control monitoring

CLI Commands

Learn all available commands and their usage patterns.

Explore CLI

Event Filtering

Control what gets monitored with filtering options.

Learn Filtering

Risk Detection

Understand how risks are identified and scored.

About Risks

Ready to Get Started?

Join thousands of developers already monitoring their AI agents with Kilometers. Set up takes less than 5 minutes.