{"version":"2.1.280","anchor":"mcp-tools-gain-an-account-memory-classification-with-redac","canonical_anchor":"mcp-tools-gain-an-account-memory-classification-with-redac","heading":"MCP tools gain an 'account memory' classification with redacted error text","tier":"internal","area":"MCP","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.280\/e\/mcp-tools-gain-an-account-memory-classification-with-redac","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.280","markdown":"### MCP tools gain an 'account memory' classification with redacted error text\n\nMCP tools tied to memory now get an 'account memory' tag, and their error messages are redacted in logs\n\n**What**\n\nMCP (Model Context Protocol, the standard Claude Code uses to talk to external tool servers) tool listings now carry a new `accountMemory` flag. It is set for:\n\n- the built-in dynamic memory server\n\n- tools with specific names such as `memory_read` and `memory_write`\n\n- tools from trusted SDK configs\n\nWhen a tool marked this way throws an error during a tool call, Claude Code no longer logs or throws the raw error text. Instead it replaces the message with a placeholder showing only the character count.\n\n**Why**\n\nMemory tools can return the contents of a user's memory file in their answers, so an error message from one of them could leak that file's contents into logs. Redacting the text keeps that private content out of logs while still surfacing that an error occurred.\n\n- Area: MCP\n- Names: `accountMemory`\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 2\/5"}