Claude Code

Anthropic's coding agent for the terminal and editors, a closed tool that talks to Anthropic's servers.

Benchmark review. Giants are reviewed rarely, to show what the tools in the directory are measured against. This one is not listed in the directory, and it is an example, not a recommendation. Scores are our opinion on the six questions, based on public sources and dated October 2026.

code.claude.com

Image from code.claude.com, taken 2026-10-30, shown to help you recognise the tool. It belongs to its maker. Ask us to remove it.

Claude Code does real work in a real project, and its docs are unusually open about where your data goes. Anthropic says that on Free, Pro and Max plans it will train new models on your data, Claude Code included, when the setting that allows it is on, with five years of retention if it is on and 30 days if it is off. Team, Enterprise and API use fall under commercial terms with no training and a standard 30 days. Usage metrics go out by default on the Claude API and can be turned off with DISABLE_TELEMETRY, and sending feedback with /feedback uploads the conversation.

Use is governed by Anthropic's Consumer or Commercial Terms and the page carries an all rights reserved notice, so it is not open source. Anyone who bundles it in a product has to run the binary unmodified, and each end user needs their own credentials. Power is a 1 because one company runs the models and sets the terms. This site was built with help from Claude, so weigh this review with that in mind. As of October 2026 this is our opinion, based on public sources, and if you spot an error, write to hello@toolsforconviviality.org and we will check it and correct it. Compare it with the tools in Coding agents.

What stands out

• A closed coding agent that sends your prompts and code to Anthropic's servers
• Consumer plans train models when the setting is on, with 5 years of retention, or 30 days when off
• Metrics are on by default and can be switched off