Elevated errors on claude.ai, Claude Code, and platform.claude.com
- ID
- 46182
- Status
- summarized
- Published
- 13 Nov 2026, 6:18 PM
- Fetched
- 13 Nov 2026, 10:48 PM
- Provider
- Hacker News
- Category
- dev-community
- Original URL
- https://status.claude.com/incidents/nn11wk485zfn
- Source URL
- https://hnrss.org/best
Summary
- Score
- 5.5
- Created
- 13 Nov 2026, 10:49 PM
- Tags
- Audience
- developersvibe_codersai_agent_userssaas_founders
What happened
An incident on 13 Nov 2026 took down authentication across claude.ai, Claude Code, platform.claude.com (Console), and Claude Cowork. Anthropic opened the investigation at 09:58 UTC, said a fix was deployed at 11:05 UTC, and marked it resolved at 11:58 UTC — roughly a two-hour window. The Hacker News thread drew 342 points and 300 comments.
Why it matters
If Claude Code or the Console API sits in your build or agent loop, this was a two-hour authentication failure with no published root cause — so the practical decision is whether your tooling degrades gracefully (queued jobs, a second provider behind an abstraction) or just stops. Anyone demoing or shipping on a single hosted model endpoint during a window like this had no local workaround.
Discussion angle
The status page gives timestamps but no cause — what would you actually need in a postmortem (auth layer, rate limiting, regional routing) before deciding whether to add a fallback provider to your agent stack?