{"version":"2.1.248","anchor":"agent-status-read-out-of-the-agent-s-own-prose","canonical_anchor":"agent-status-read-out-of-the-agent-s-own-prose","heading":"Agent status read out of the agent's own prose","tier":"notice","area":"Agents","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248\/e\/agent-status-read-out-of-the-agent-s-own-prose","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248","markdown":"### Agent status read out of the agent's own prose\n\nAgent status is read from their own text, recognising phrases like \"needs input:\" and \"blocked:\".\n\n**What**\n\nBackground and managed agents are now classified by scanning their output for state lines. Recognised prefixes are `needs input:`, `failed:`, `blocked:` and `I'm blocked:`, each accepting a colon, em dash or en dash, and a separate detector matches \"Giving up\" and \"The task is not actionable\" to mark an agent as having stopped.\n\n**Details**\n\n- Matching is case-insensitive and captures 3 to 200 characters of the reason up to the end of the line.\n\n**Evidence**\n\n`\/(?:^|\\n)\\s*needs input\\s*[:\\u2014\\u2013-]\\s*(.{3,200}?)(?=\\n|$)\/gi`\n\n- Area: Agents\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 4\/5"}