I recorded 87 Claude Code sessions to find out where the money goes
I wanted to know why AI coding agents get expensive on real codebases, so I stopped guessing and started recording. Every session in this post is a real Claude Code run (Claude Sonnet 5), saved as a full transcript with the cost Claude Code reported. 87 sessions in total, on two open-source repositories. Along the way I built knock , an open-source MCP server that gives the agent a map of the…
A reporter spent the past few weeks recording Claude Code sessions to uncover the costs of AI coding agents on real codebases. The reporter ran 87 sessions on two open-source repositories, building an open-source MCP server named "knock" that provides the agent with a map of the codebase. The reporter found that using knock reduced costs by 40% compared to not using knock, with a median cost of $0.25 versus $0.42 for Claude Code alone.
The key factors that contributed to the cost savings were reduced tool calls, faster model time, and the ability for Claude Code to skip searching and directly access the relevant code sections.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
