{"version":"2.1.292","anchor":"at-mentioning-a-file-that-is-too-large-now-gives-a-read-in-po","canonical_anchor":"at-mentioning-a-file-that-is-too-large-now-gives-a-read-in-po","heading":"@-mentioning a file that is too large now tells Claude to read it in parts","tier":"notice","area":"Attachments","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292\/e\/at-mentioning-a-file-that-is-too-large-now-gives-a-read-in-po","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292","markdown":"### @-mentioning a file that is too large now tells Claude to read it in parts\n\nWhen you @-mention a file too large to attach, Claude is told its size and to read it in portions\n\n**What**\n\nWhen you mention a file with `@` in your prompt and the file is too large to attach whole, Claude Code now handles that case on its own. The model is told:\n\n- that the file's contents were not attached because it is too large to read at once\n\n- how big the file is\n\n- that reading it without a limit will fail, and that it should read it in portions using an offset (where to start) and a limit (how much to read)\n\nWhen listing a mentioned folder's contents fails, that is now reported separately as well.\n\n**Why**\n\nInstead of trying to read the whole file and failing, Claude gets clear instructions for reading a large file piece by piece.\n\n- Area: Attachments\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}