How are AI agents used? Evidence from 177,000 MCP tools

TL;DR

This study analyzes over 177,000 AI agent tools to understand their usage patterns, categorization, and implications for task automation and regulation, revealing a significant rise in action tools over time.

Contribution

It provides a comprehensive empirical analysis of AI agent tools from MCP repositories, categorizing their functions and assessing their growth and potential regulatory implications.

Findings

Software development dominates tool usage (67%)

Action tools increased from 27% to 65% over 16 months

High-stakes action tools like financial transactions are present

Abstract

Today's AI agents are built on large language models (LLMs) equipped with tools to access and modify external environments, such as corporate file systems, API-accessible platforms and websites. AI agents offer the promise of automating computer-based tasks across the economy. However, developers, researchers and governments lack an understanding of how AI agents are currently being used, and for what kinds of (consequential) tasks. To address this gap, we evaluated 177,436 agent tools created from 11/2024 to 02/2026 by monitoring public Model Context Protocol (MCP) server repositories, the current predominant standard for agent tools. We categorise tools according to their direct impact: perception tools to access and read data, reasoning tools to analyse data or concepts, and action tools to directly modify external environments, like file editing, sending emails or steering drones in the physical world. We use O*NET mapping to identify each tool's task domain and consequentiality. Software development accounts for 67% of all agent tools, and 90% of MCP server downloads. Notably, the share of 'action' tools rose from 27% to 65% of total usage over the 16-month period sampled. While most action tools support medium-stakes tasks like editing files, there are action tools for higher-stakes tasks like financial transactions. Using agentic financial transactions as an example, we demonstrate how governments and regulators can use this monitoring method to extend oversight beyond model outputs to the tool layer to monitor risks of agent deployment.