From grok
Ask Grok a one-shot question, leaning on live web/X search (the headline differentiator)
How this command is triggered — by the user, by Claude, or both
Slash command
/grok:ask [--model <slug>] [--no-search] [--search] [--background] <question>This command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
Ask Grok a one-shot question. Raw slash-command arguments: `$ARGUMENTS` Execution mode: - If the raw arguments include `--background`, use the background flow. - Otherwise use the foreground flow. Foreground flow: - Return the command stdout verbatim — do not summarize, paraphrase, or add commentary before or after it. - Preserve any inline markdown citation links Grok includes. Background flow: - Launch with `Bash` in the background: - Do not wait for completion in this turn. The command prints a job id immediately; tell the user: "Grok ask started in the background. Check `/grok:sta...
Ask Grok a one-shot question.
Raw slash-command arguments:
$ARGUMENTS
Execution mode:
--background, use the background flow.Foreground flow:
node "${CLAUDE_PLUGIN_ROOT}/scripts/grok.mjs" ask "$ARGUMENTS"
Background flow:
Bash in the background:Bash({
command: `node "${CLAUDE_PLUGIN_ROOT}/scripts/grok.mjs" ask "$ARGUMENTS"`,
description: "Grok ask",
run_in_background: true
})
/grok:status for progress, then /grok:result <job-id> for the answer."Notes:
web_search config setting. Pass --no-search to disable it for one call, or --search to force it on.search_model (grok-build), which searches reliably. Override per call with --model <slug>.npx claudepluginhub trevoraspencer/grok-plugin-cc --plugin grok/askProvides structured architectural guidance on technical questions via consultation with systems design, technology strategy, scalability, and risk experts, including analysis, recommendations, and next steps.
/askQueries multiple AI agents (Gemini, OpenAI, Grok, Perplexity) for diverse perspectives on architecture decisions, technology choices, debugging dead-ends, and security tradeoffs.
/askAsks NotebookLM a question about the current notebook's content via `notebooklm ask` CLI, displaying the response with optional source references via --json.