{"version":"2.1.248","anchor":"new-guardrails-in-the-model-s-instructions","canonical_anchor":"new-guardrails-in-the-model-s-instructions","heading":"New guardrails in the model's instructions","tier":"notice","area":"System Prompt","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248\/e\/new-guardrails-in-the-model-s-instructions","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248","markdown":"### New guardrails in the model's instructions\n\nClaude is now told not to fake clicks, touch shell startup files, or call tools during summaries.\n\n**What**\n\nFour instruction fragments were added: the model must not use AppleScript, System Events or shell commands to fake clicks and keystrokes, must not save files into shell startup files, ssh keys or launch agents, must not claim it is editing an artifact page, and must not make tool calls during an analysis or summary step.\n\n**Details**\n\n- These are prompt text, so they shape behaviour rather than hard-blocking the actions.\n\n**Evidence**\n\n`System Events, shell commands, or any other method to send clicks or keystrokes to this app.`\n\n- Area: System Prompt\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 4\/5"}