I gave Claude Code, Codex, and Cursor the same memory, and stopped repeating myself

The author of a series of reports has been testing AI tools, including Claude, Codex, and Cursor, by giving them the same memory and avoiding repetition. They have been bouncing between different AI tools for a long time, and this has become a normal part of their workflow, according to XDA Developers.

The author tested Claude by assigning it different senior engineering roles, including full-stack engineer, debugging engineer, frontend engineer, and performance engineer, reports Tom's Guide. Claude reviewed an expense tracker app four times, each time concentrating on a different set of problems and occasionally finding issues that the previous "developer" had missed.

According to Dev.to, the author also measured the cost of installing 107 skills, 38 agents, and 15 commands into Claude Code, which was 9,857 tokens. This cost is paid for every session, even if the skills are not used. Techmeme reports that Anthropic will permanently raise weekly Claude Code limits by 25% on September 14 for most plans.

Brief written by urgent.news from XDA Developers, Tom's Guide, Dev.to, Techmeme — 4 reports on this story. Machine-written — may contain errors; check the original before relying on it.