{"version":"2.1.285","anchor":"memory-extraction-mcp-mode-tracks-unfinished-passes-and-re","canonical_anchor":"memory-extraction-mcp-mode-tracks-unfinished-passes-and-re","heading":"Memory extraction retries turns it did not finish reviewing","tier":"notice","area":"Memory","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.285\/e\/memory-extraction-mcp-mode-tracks-unfinished-passes-and-re","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.285","markdown":"### Memory extraction retries turns it did not finish reviewing\n\nWhen a pass that saves memories from your conversation stops early, the next pass looks at the same turns again\n\n**Unclear.** It is not clear which setups use the MCP mode of memory extraction.\n\n**What**\n\nClaude Code can review your conversation and save useful points as memories. It uses a bookmark to track which turns (one message from you and Claude's reply) it has already looked at. In the mode that runs through MCP, the Model Context Protocol for connecting outside tools, that bookmark used to move forward after every pass, whatever happened. Now it stays put when:\n\n- a pass stops early, or\n\n- the pass's saved changes are refused because a newer message has arrived.\n\nThe next pass then reviews those same turns again. After several unfinished passes in a row, Claude Code gives up and moves the bookmark on. Usage reports now also record how many memories were written, how many were refused, and whether the retry limit was reached.\n\n**Why**\n\nMemories are less likely to be silently skipped when a pass fails.\n\n- Flag `tengu_bramble_lintel`: Not enough to say (read for one account on one subscription tier against v2.1.285; this account: no value returned, anonymous baseline: no value returned, compiled default: not a boolean we can read) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.\n- Area: Memory\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 3\/5"}