The Pulse
Anthropic resolves Claude errors after API and Code disruption
Anthropic marked an incident affecting Claude’s website, API and coding tool resolved on November 13. Minutes earlier, the company had reported delays to scheduled tasks and security features.

Errors affecting Claude’s website, API and coding tool have been resolved after a disruption that also reached connectors and other services, according to Anthropic’s status page. The company marked the incident resolved at 15:38 UTC on November 13, shortly after reporting that some features were still delayed or recovering.
The incident followed a separate authentication problem earlier the same day. That earlier issue affected Claude’s website and coding tool, as well as the Console and Cowork services.
Errors across services
At 13:09 UTC, Anthropic said it was investigating reports of degraded performance on claude.ai, the Claude API and Claude Code. An update at 13:52 UTC added connectors, including Claude Docs and Claude Sheets, to the affected services.
By 14:03 UTC, the investigation also covered Claude Security and Claude Managed Agents. The incident record lists Claude Cowork among the affected services.
At 14:36 UTC, the company reported recovery on claude.ai and the API, while cautioning that errors could return. Connectors and connector use in Claude Code remained degraded, as did Claude in Meetings, Managed Agents and Security. At that point, mobile apps were affected in the same way as the website.
Recovery and remaining delays
In a 15:33 UTC monitoring update, Anthropic said error levels had returned to normal on claude.ai, its mobile apps, the API and Claude Code. Existing connectors, including Docs and Sheets, were working again, while Managed Agents and Claude Meetings were recovering.
Adding a new connector on claude.ai was temporarily unavailable, the company said at that point. Scheduled tasks and routines could run more than two hours late, and some runs had been skipped. Claude Security scans, reviews and project setup remained delayed.
The incident was marked resolved five minutes later. StatusStack reported that, at about 10:50 a.m. EST, Claude’s status dashboard still listed claude.ai and Cowork as having degraded performance, while the API, Code and Console were listed as operational.
Earlier authentication issue
Earlier on November 13, Anthropic had investigated a separate problem with users authenticating to claude.ai and Claude Code. The company said it deployed a fix at 11:05 UTC and was monitoring recovery; it marked that incident resolved at 11:58 UTC.
The earlier incident record also lists Claude Console and Claude Cowork as affected. StatusStack reported that the Claude API remained operational during that incident, unlike in the later disruption.
Separately, StatusStack reported that an incident affecting the Microsoft 365 connector remained open. Anthropic moved that issue to monitoring at 10:40 a.m. EST on November 13 and reported signs of recovery, according to the service-monitoring site.
Sources
Companies in this story
Related stories
- Anthropic Confirms Claude Code Rule Against Recreating Specific Games
- Anthropic adds experimental Code Mode to Claude Code
- Apiiro Finds Claude Code Approval Files Leaking Secrets on GitHub
- Anthropic advises deleting "think hard" lines from Claude Opus 5.5 prompts
- Anthropic opens self-hosted Claude Managed Agents to private beta