Claude Code Accused of Leaking User Emails
A GitHub user alleges Claude Code sent their email address in HTTP headers, prompting a privacy module and a planned change from Anthropic.
Updated on Nov. 6, 2026 in Artificial Intelligence
Live Poll
Do you trust AI coding tools to keep your personal data private?
On November 5, 2026, a GitHub user filed issue #107757 alleging that Claude Code transmitted their email address through HTTP User-Agent headers, and says they now receive dozens of scam emails per day. Claude Code versions 2.1.317 and later include a privacy module called cc-plugin-privacy designed to block this behavior.
Why it matters
The report raises questions about how AI coding tools handle personal account data when making outbound web requests. The public evidence does not establish that the alleged exposure caused the scam emails or affected every user.
The evidence is a single user-filed GitHub issue, which reports dozens of scam emails per day but includes no raw HTTP request, sample header, or reproduction steps. It is unclear whether the alleged exposure caused the scam emails or affects all users, and the issue was still open at review.
The players
Anthropic
Anthropic is the AI company that develops Claude and Claude Code, and a quoted employee said it plans to remove account emails from model context in an upcoming release.
The details
The reporter says Claude Code placed their email address in HTTP User-Agent headers, which reach destination websites with each request. A quoted explanation in the thread describes Claude using an email available in its context as a contact address when writing a curl request. A built-in module, cc-plugin-privacy, is enabled by default in version 2.1.317 and later and can be disabled through /plugins, though some commenters could not locate it.
Timeline
November 5, 2026: Reporter files GitHub issue #107757 alleging the email leak.
Roadmap
The alleged leak arises as AI coding agents increasingly act on users' behalf with web requests, making privacy controls like Claude Code's cc-plugin-privacy module a point of focus. Anthropic's planned removal of account emails from model context signals a shift in how such tools handle personal data.
Users of Claude Code who have not updated to version 2.1.317 or later may want to check whether their account email could appear in outbound requests. The cc-plugin-privacy module is enabled by default in newer versions and can be disabled through /plugins if users choose.
The takeaway
Developers and users of AI coding tools should review what personal data the tools can send over the network, and should confirm that privacy modules are enabled. Vendor promises to change behavior in future releases do not address current exposure.
Further reading
For more on the tools behind this report, see Artificial Intelligence coverage.
Source note: This article includes information reported by Cyber Security News.
Live Poll
Do you trust AI coding tools to keep your personal data private?
