Designing the hf CLI as an agent-optimized way to work with the Hub
Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated.
Hugging Face has redesigned its official hf command-line tool to serve both human users and AI coding agents like Claude Code, Codex, and Cursor, automatically detecting when an agent is driving it and adjusting output accordingly. The CLI now tags and tracks agent traffic, which has grown significantly since tracking began in April 2026, with Claude Code alone accounting for roughly 40,000 users and nearly 49 million requests. Benchmarking showed that on complex, multi-step tasks, agents using non-CLI methods like curl or the Python SDK consumed up to six times as many tokens as those using the optimized hf CLI.
The Hugging Face hf CLI tool has been redesigned to better serve both human users and AI coding agents, optimizing outputs differently for each. AI agents receive compact, structured responses to minimize token usage, while humans get more visually rich terminal outputs. Early data shows significant adoption by coding agents like Claude Code and Codex, with the CLI proving up to 6 times more efficient than manual approaches for complex tasks.