{"version":"2.1.283","anchor":"per-model-weekly-usage-rows-no-longer-crash-on-a-non-array-l","canonical_anchor":"per-model-weekly-usage-rows-no-longer-crash-on-a-non-array-l","heading":"Weekly per-model usage rows handle unexpected server data","tier":"notice","area":"Usage & Limits","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.283\/e\/per-model-weekly-usage-rows-no-longer-crash-on-a-non-array-l","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.283","markdown":"### Weekly per-model usage rows handle unexpected server data\n\nThe \"Current week (model)\" usage rows now show nothing instead of failing when the server sends limits in an unexpected shape\n\n**What**\n\nClaude Code can show a \"Current week ()\" usage row for each model. The code that builds those rows now returns an empty list when the limits data from the server is not a list. Before, it only handled the case where the data was missing entirely.\n\n**Why**\n\nIf the server returns the limits in an unexpected shape, the usage display no longer breaks. It simply shows no per-model weekly rows.\n\n- Area: Usage & Limits\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5"}