{"version":"2.1.319","anchor":"permission-prompt-tool-different-handling-for-client-requ","canonical_anchor":"permission-prompt-tool-different-handling-for-client-requ","heading":"Permission prompt tool hands some requests back to the default handler","tier":"notice","area":"Permissions","scope":"individual","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.319\/e\/permission-prompt-tool-different-handling-for-client-requ","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.319","markdown":"### Permission prompt tool hands some requests back to the default handler\n\n`--permission-prompt-tool` now uses the default handler when a request carries a client tool-use id and the tool has no input schema\n\n**Unclear.** It is not clear which requests carry a client-supplied tool-use id, so when the default handler takes over.\n\n**What**\n\n`--permission-prompt-tool` names an MCP tool that answers permission prompts when Claude Code runs without you watching. Some requests now skip it and go to Claude Code's default permission handling instead. This happens when both of these are true:\n\n- the request carries a tool-use id supplied by the client\n\n- the MCP tool has no input schema, the description of what input it accepts\n\nThe check for whether the prompt tool's server has gone away now has three outcomes instead of two.\n\n**Why**\n\nThis changes how permission prompts are routed for programs that drive Claude Code through the SDK.\n\n- Area: Permissions\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: yes"}