
Share
GitHub’s latest update automatically tracks user interactions with its CLI, sparking debate over privacy and consent in an era where data collection becomes increasingly pervasive.
GitHub has quietly rolled out a new feature that enables client-side telemetry collection by default in its command-line interface (CLI). The update, which began with the v2.91.0 release on April 22, 2026, aims to gather usage data to improve the CLI and better understand how features are utilized, particularly as AI agents become more prevalent.
If you prefer not to share your usage data with GitHub, you can opt out by running the following command:
gh config set analytics opted-out true

This will disable telemetry collection for your CLI sessions.
While the addition of telemetry is a common practice in software development, enabling it by default without clear communication has raised concerns among privacy-conscious developers. The lack of a detailed list of data points collected adds to these concerns.
A GitHub spokesperson provided the following statement:
"GitHub CLI now includes client-side usage telemetry, beginning with the v2.91.0 release today, to better understand how developers use the tool across both interactive and increasingly agentic workflows. Telemetry is enabled by default with a clear opt-out. These insights will help the team identify friction points and prioritize the features that matter most, particularly as usage patterns evolve with the rise of AI agents. GitHub CLI is open source, so you can review the telemetry implementation in the cli/cli repository."
Enabling telemetry by default without a clear announcement or detailed documentation has sparked a debate within the developer community. While data collection can provide valuable insights for product improvement, it's crucial to maintain transparency and respect user privacy. If you're concerned about your data, consider opting out and staying informed about future updates.
Tags
Original Sources
About the author
Kai built ML infrastructure at a Bay Area startup before developing an obsession with transformer architectures and inference optimisation that eventually pulled him out of product work entirely. A stint at a compute research lab sharpened his instinct for what actually matters in a model release versus what is marketing. He writes from the inside — from the perspective of someone who has debugged the systems he is describing at three in the morning. He is allergic to hype and instinctively drawn to the unglamorous plumbing questions that everyone else skips over.
More from The Engineer →This Week's Edition
25 April 2026
133 articles
Related Articles
Related Articles
More Stories