메뉴
Ask xAI Grok a one-off question with optional live web search. Use when the user wants Grok's opinion or a quick answer that benefits from Grok's distinct training (recent web/X awareness, code-focused models). Read-only by default. Returns Grok's stdout verbatim.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Run a multi-angle Grok analysis in a single call — Grok analyzes the task from several expert angles (security, review, research, tests) and synthesizes one consolidated answer with a section per angle plus a verdict. Use when the user wants a thorough, multi-perspective audit/review/analysis of something ("find everything wrong with…", "review this from all angles", "deep multi-angle look at…") that benefits from more coverage than a single-shot question. Read-only by default.
Deep research with xAI Grok using effort=max, live web search, and Grok's self-verification --check loop. Use when the user wants a thorough investigation of a topic, recent events, or claims that need source-grounding. Returns Grok's research output verbatim.
Delegate a long-form coding task, root-cause investigation, multi-file refactor, or open-ended fix to xAI Grok via the companion's `task` subcommand. Use when the work is too large or specialized to handle inline and benefits from Grok's distinct training (live web search, code-focused models, 512K context).
Generate an image from a text description using xAI Grok's `/imagine` builtin. Use when the user asks for an image, picture, illustration, logo, mockup, or visual artifact. Returns the absolute path to the saved image plus an `open` command to view it.
Internal helper contract for calling the grok-companion runtime from Claude Code
Internal guidance for composing Grok prompts for coding, review, diagnosis, and research tasks inside the Grok Claude Code plugin
| name | grok-ask |
| description | Ask xAI Grok a one-off question with optional live web search. Use when the user wants Grok's opinion or a quick answer that benefits from Grok's distinct training (recent web/X awareness, code-focused models). Read-only by default. Returns Grok's stdout verbatim. |
Use this skill when the user asks Grok a question directly, or when you need a second opinion / cross-check from a different model than the one you're running in. The companion script wraps the local grok CLI with environment scrubbing, ANSI sanitization, capability probing, and prompt-injection defenses.
Do not use this skill for:
grok-rescue skill instead.grok-imagine.--effort max + --check self-verification → use grok-research.Run exactly one bash command:
node "${CLAUDE_PLUGIN_ROOT}/scripts/companion.mjs" ask "<the user's question, properly quoted>"
The CLAUDE_PLUGIN_ROOT env var is set by both Claude Code and Codex CLI. The companion accepts the full argument string and parses its own flags.
--model <name> — override default Grok model. Available models (run /grok:models): grok-build (default), grok-composer-2.5-fast.--effort <low|medium|high|xhigh|max> — reasoning effort. Note: the current default model (grok-build) declares no reasoning-effort support, so the companion silently strips --effort (with a stderr warning) to avoid a 400 from upstream. It only takes effect on a model that advertises supports_reasoning_effort.--timeout <duration> — e.g., 90s, 5m, 0 to disable. Default unbounded.--no-web-search — disable Grok's live web search (on by default).--check — append Grok's self-verification loop to the answer.--rules <text|@file> — extra system rules.--sandbox <profile> — restrict tool/filesystem access.[hint: ...] line about authentication, surface it and tell the user to run grok login or set XAI_API_KEY.