GitHub issue #74260 in anthropics/claude-code, opened 2026-07-04 by federbenjamin, records a report that on Claude Code 2.1.201 with adaptive thinking, assistant text emitted mid-turn — a text block followed by another thinking block or further tool calls in the same turn — is frequently neither rendered on any surface nor persisted to the session transcript JSONL; in one measured session, 12 of 13 mid-turn status texts vanished from both the display and the JSONL.
Verbatim Excerpts from Cited Records
“**Environment:** Claude Code 2.1.201, macOS (darwin 25.5), model `claude-fable-5` (interleaved/adaptive thinking), terminal TUI (also reproduced in a claude.ai/code bridged child session, `CLAUDE_CODE_CHILD_SESSION=1`).”
·
anthropics/claude-code (issue 74260 opening body, Environment section)
·
2026-07-04T19:26:54Z
·GitHub Issue ↗
“**Summary:** Assistant text emitted mid-turn — a `text` block followed by another `thinking` block and/or further tool calls in the same turn — is frequently neither rendered on any surface nor persisted to the session transcript JSONL.”
·
anthropics/claude-code (issue 74260 opening body, Summary section)
·
2026-07-04T19:26:54Z
·GitHub Issue ↗
“In one measured session, 12 of 13 mid-turn status texts vanished from both the display and the JSONL.”
·
anthropics/claude-code (issue 74260 opening body, Impact section)
·
2026-07-04T19:26:54Z
·GitHub Issue ↗
Counter-record: Issue remains open as of 2026-10-23 with 34 comments. The reporter distinguishes this from #23862 / #58207 (AskUserQuestion dialog overlay hiding preceding text): here the text is absent from the transcript JSONL itself, placing the loss upstream of rendering. Related but distinct from C015 (#87640, a safeguard refusing a message): #74260 concerns mid-turn text blocks silently dropped, not refusals. No maintainer resolution or author retraction appears in the thread; the reporter offered session JSONL evidence on request.
Falsification probe: If issue #74260 is deleted without archive, if the author edits the opening body so the quoted sentences no longer match verbatim, or if the author/maintainers post a definitive retraction showing the described behavior was mischaracterized, this claim drops to Withdrawn.
Audit verification: Verified 100% continuous verbatim substring against official GitHub REST API by
scripts/verify_claim_verbatim.py.