{"version":"2.1.321","anchor":"new-privacy-hooks-module-refuses-shell-calls-pairing-the-a","canonical_anchor":"new-privacy-hooks-module-refuses-shell-calls-pairing-the-a","heading":"Shell commands pairing your account email with a User-Agent are refused","tier":"notice","area":"Privacy","scope":"individual","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321\/e\/new-privacy-hooks-module-refuses-shell-calls-pairing-the-a","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.321","markdown":"### Shell commands pairing your account email with a User-Agent are refused\n\nA built-in privacy hook, on by default, refuses shell commands that contain your account email and also set a User-Agent\n\n**Unclear.** What decides whether this module is available on a given setup is not stated.\n\n**What**\n\nClaude Code now includes a built-in privacy hooks module. Hooks are small programs that run automatically during a session. This one is on by default, and its on or off setting is honoured only from trusted settings.\n\nIt refuses a shell command that does both of these:\n\n- Contains the email address of the account you are signed in with\n\n- Sets a User-Agent, the label a program sends to identify itself to a website, using flags such as `curl -A` or `wget -U`, or similar ones for `iwr`\n\nThe refusal message tells Claude to resend the command without the email address, or to split it into separate calls.\n\n**Why**\n\nIt stops your email address and a browser-style identity being sent together in one request. Because it is on by default and works quietly, a command Claude runs may be refused without you expecting it.\n\n- Area: Privacy\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: yes"}