{"version":"2.1.301","anchor":"slash-command-in-prompt-check-and-projects-relay-counting","canonical_anchor":"slash-command-in-prompt-check-and-projects-relay-counting","heading":"Unknown slash commands are detected, and relayed prompts counted","tier":"notice","area":"Elsewhere","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.301\/e\/slash-command-in-prompt-check-and-projects-relay-counting","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.301","markdown":"### Unknown slash commands are detected, and relayed prompts counted\n\nClaude Code can now tell when a prompt is a slash command it does not know, and counts relayed prompts separately\n\n**What**\n\nA slash command is input starting with `\/` that runs a command instead of going to Claude. Claude Code now checks whether a prompt looks like a slash command that is not in its list of known commands. The check says no:\n\n- in \"poll-event\" mode\n\n- when `skipSlashCommands` is set, unless the prompt was scheduled by the model or typed through the bridge\n\nSeparately, Claude Code's prompt counts now track relayed prompts in their own group.\n\n**Why**\n\nThis changes how Claude Code sorts what you type when deciding whether to treat it as a command, and how prompts are counted.\n\n- Area: Elsewhere\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}