{"version":"2.1.282","anchor":"post-turn-memory-config-gains-extract-transport-tools-o","canonical_anchor":"post-turn-memory-config-gains-extract-transport-tools-o","heading":"Post-turn memory config accepts an extract_transport choice","tier":"use","area":"Memory","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.282\/e\/post-turn-memory-config-gains-extract-transport-tools-o","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.282","markdown":"### Post-turn memory config accepts an extract_transport choice\n\nThe post-turn memory config now takes an extract_transport option, \"tools\" or \"text\", defaulting to \"tools\"\n\n**Unclear.** What the \"text\" setting actually changes in memory extraction is not known.\n\n**What**\n\nPost-turn memory is an optional feature that is only active when the environment variable `CLAUDE_CODE_POST_TURN_MEMORY` is set. Its settings are given as JSON in `CLAUDE_CODE_POST_TURN_MEMORY_CONFIG`. That JSON now accepts a key `extract_transport` with one of two values:\n\n- `\"tools\"`, the default\n\n- `\"text\"`\n\n**Why**\n\nThis suggests memory extraction can now run with the model writing plain text, rather than making tool calls (structured requests to Claude Code's built-in tools).\n\n- Area: Memory\n- Names: `CLAUDE_CODE_POST_TURN_MEMORY_CONFIG`, `CLAUDE_CODE_POST_TURN_MEMORY`, `extract_transport`\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 4\/5"}