Developers on Hacker News reported that Claude Code became inaccessible for an hour, yet the official status page remained silent. Users shared Reddit links showing daily‑expiring OAuth keys, 500 errors, and timeouts when attempting re‑authorization. The incident coincided with a surge in API demand from enterprise pilots, leaving many questioning the service’s reliability during critical coding sessions.

Anthropic’s engineering team eventually updated the page to acknowledge elevated error rates, prompting brief applause from the community. Critics argue that a robust status system should tie directly to internal health checks rather than rely on ad‑hoc updates. Some users suggested third‑party monitors that continuously probe endpoints, providing transparent latency and capacity data similar to traditional infrastructure dashboards.

Beyond the outage, developers highlighted deeper integration gaps: Claude’s CLI still lacks full MCP support, and open issues in its repository exceed 5,000, many flagging regressions. Comparisons with self‑hosted models like Qwen show lower token costs and comparable speed, nudging some teams toward local alternatives until Anthropic resolves its operational transparency shortcomings. Such gaps risk eroding developer trust and may push teams toward AI tools.