Claude Code v2.1.282 fixes thinking-block loss, compaction failures and 400 errors
ashwin-ant · ghdev · 2026-09-25
Claude Code v2.1.282 ships a batch of fixes and settings: a maxProseWidth option for wide terminals, telemetry notice via /status and claude doctor, allowClaudeInChromeWithManagedMcp for Chrome + exclusive managed-mcp, and a gateway readiness grace window for Postgres failovers.
Key fixes: 400 errors from undecryptable web-search history; --continue/--resume re-sending altered messages and dropping prior reasoning; extended thinking lost on slash commands or --tools relaunch; "Invalid data in redactedthinking block" (now retries once after dropping thinking blocks); compaction failing on refused summarization (retries on fallback model); and errors after safety-related model switches with thinking off.
More from coding & agent
- AI compliance startup CompAI hits 1,000+ paying business customers in under a year — JosephJacks_ · 2026-09-25
- Sleep Data for Personal AI? Separating Daily Signals from Hard Rules — sujingshen · 2026-09-25
- Markdown Is the New Source Code: How to Manage Runbooks, Prompts and Rules for Agents — arpit_bhayani · 2026-09-25
- The personal agent supercycle needs hard authorization boundaries, not autonomy — sujingshen · 2026-09-25
- Mnemos.Field nears launch: a virtual world where humans and AI agents both register and participate — RileyRalmuto · 2026-09-25
- SemIf open-sources a Jev-style interface: typed option probabilities from a 4B model, no JSON parsing — JeremyCMorgan · 2026-09-25