Claude Code is locking people out for hours
agents anthropic claude
| Source: HN | Original article
Anthropic’s Claude Code has begun cutting off users for several hours after they exhaust the service’s daily quota, sparking frustration across developer forums. The terminal‑based coding assistant, which blends large‑language‑model reasoning with IDE‑style actions, enforces a hard limit of roughly 40 short exchanges on its free tier. Once that ceiling is reached, the platform returns a “rate‑limit exceeded” error and refuses further requests until the quota resets, a window that can stretch to four or six hours depending on the user’s region.
The outage matters because Claude Code has quickly become a go‑to tool for rapid prototyping and code‑base navigation, especially among Nordic startups that favour open‑source‑friendly AI. Extended lockouts erode the productivity gains the agent promises and may push developers toward competing offerings such as GitHub Copilot, OpenAI’s Assistants API, or community‑built alternatives like the “Desktop Pet” Copilot we covered earlier. Moreover, the incident highlights a broader tension in the AI‑as‑a‑service market: providers must balance generous free access that fuels adoption with sustainable cost structures, and Anthropic’s current throttling strategy appears to tip the scale toward revenue protection at the expense of user experience.
We first flagged Claude Code’s rate‑limit quirks in our “Stop hitting Claude rate limits mid‑session” guide (April 7, 2026). Since then, the lockout duration has lengthened, suggesting a policy shift rather than a temporary glitch. Developers are already experimenting with multi‑provider wrappers and CLI tricks to spread requests across OpenAI and Anthropic models, but such workarounds add complexity.
What to watch next: an official statement from Anthropic clarifying whether the extended limits are permanent, a possible revision of the free‑tier quota, and any upcoming feature roll‑outs that might include more granular usage dashboards. Competitors’ responses and the emergence of community‑driven proxy tools will also indicate whether Claude Code can retain its foothold in the fast‑moving AI‑coding arena.
Sources
Back to AIPULSEN