Xai Research

About

xai-research is an agent skill that sends research queries to xAI Grok through the llmCLI OAuth forwarder, so developers can pull live web results, X posts, or conversational Grok answers from inside their usual coding agent. It is aimed at developers who already use llmCLI-compatible agents and want Grok’s search surfaces without maintaining separate xAI credentials in every project. Reach for it when you need fast external signal—competitor pages, launch chatter on X, or a structured answer you can paste into a spec or validation doc—not when you are implementing product code. Because it is an integration rather than a methodology skill, you stay in control of prompts and how results land in your repo; the skill’s value is reliable, authenticated access to Grok-backed research from agent sessions. Pair it with planning or documentation skills after you have enough research to narrow scope.

• Routes research queries to xAI Grok via an llmCLI OAuth forwarder—no manual API key plumbing in the agent session.
• Supports three modes: web search, X (Twitter) search, and plain Grok chat for follow-up synthesis.
• Fits agent workflows where you want cited, up-to-date answers without leaving Claude Code, Cursor, or Codex.
• OAuth-forwarded access keeps secrets out of prompts while still calling Grok-backed search endpoints.
• Composable with other skills: use outputs as input for scoping, landing copy, or SEO briefs in later phases.