5 questions · ~2 minutes · browser-only, no signup, no telemetry · CC0 · v1.0 (2026-05-30)
The wedge: When extended thinking is enabled, a session can hit API Error: 400 messages.N.content.M: thinking or redacted_thinking blocks in the latest assistant message cannot be modified and never recover within that session. The corrupted message persists in the transcript; every subsequent turn re-sends and re-fails identically. Documented across 15+ open issues (central case #63147, 33 reactions) — reproduces on v2.1.143 through v2.1.156. The cluster has four sub-patterns with different triggers and different recoveries. Misapplied /clear burns context on 13B/C/D when a transcript edit could have preserved it.
New as of 2026-05-29: Under /loop or other autonomous-resume harnesses, a one-time 400 becomes an unrecoverable infinite loop — the loop queues another continuation, the latest assistant message is unchanged on disk, the next request hits the same 400, the loop replays. @LMS927369 reported this on the #63147 thread. Question 5 below detects this amplification.
Disclosure: The prevention section links to two SessionStart hooks in this repository (PR #445 advisory + PR #469 opt-in BLOCK for autonomous run) and the published field guide. The hooks detect the 13A precursor before resume actually fires the 400. Free, MIT-licensed. No signup, no telemetry, no analytics beacons — this page is a single self-contained HTML file you can save and run offline.
5 Questions
Diagnosis
Top likely sub-pattern
Autonomous-run amplification detected
Your session runs under autonomous resume. The wedge becomes an unrecoverable infinite loop — every loop iteration re-fires the 400. Stop the loop now before it burns API tokens / Pool 2 quota / Max-plan rate-limit budget on requests that cannot succeed.
Immediate: Kill the loop process. The wedged transcript will not heal itself.
Then: Apply the recovery path below (typically the JSONL strip) on the transcript before any re-entry.
Then: Install extended-thinking-loop-guard.sh (PR #469 below) before resuming autonomous run.
PR #445 (cc-safe-setup, shipped 2026-05-29): extended-thinking-resume-warning.sh — SessionStart hook that detects the 13A precursor and emits a non-blocking advisory before resume fires the 400. 54 tests.
PR #469 (cc-safe-setup, shipped 2026-05-30): extended-thinking-loop-guard.sh — opt-in BLOCKING SessionStart hook for autonomous-run users. Same detection as PR #445 but exits 2 with decision JSON when armed, breaking the /loop retry cycle. 49 tests.
This page covers Cluster 13 specifically. If extended thinking failures are a recurring tax on your operation — burning quota, killing sessions, derailing long-horizon runs — the published material extends to:
cc-safe-setup ships ~800 hooks for Claude Code, MIT-licensed, 1,330 14-day-unique users. Free.
Token Book ¥2,500 includes Chapter 13 on the Cluster 13 wedge specifically — recovery, prevention, and the token economics of mid-session failure. First two chapters are free preview.
Safety Lab ¥500/month — monthly issue treating each emerging cluster with the shipped hooks, operator checklist, and upstream-status tracking. Cluster 13 is the February 2027 lead chapter.