Claude Code adds a warning before tool calls leak your account email
Version 2.1.317, shipped 3 November, asks for confirmation before a web or shell tool call sends your account email address to the network, and adds a cost report API for gateway admins.
Anthropic shipped Claude Code 2.1.317 on 3 November. The release adds a check that asks for confirmation before a web or shell tool call sends your account email address to the network, closing a path by which the tool could leak that email with nothing shown in the UI.
Claude Code already gates file and shell access behind permission prompts, but a call to an MCP server, a web fetch, or a shell command could previously carry your account email as a side effect with no separate warning. The same release adds a cost report to the Claude apps gateway admin API, letting admins read past spend by date range, developer, day or IdP group:
GET /v1/organizations/analytics/costIt also keeps a daily record of each developer’s IdP groups, retained for admin.identity_retention_days, which that cost report groups by.
Why it matters: Claude Code is routinely pointed at MCP servers and scripts you don’t fully control, and an account email is exactly the kind of identifier you don’t want leaving your machine without you noticing. If you run Claude Code against third-party MCP servers, this release is worth taking for that check alone, separately from the usual bug fixes.