Chat with Grok models via xAI API. Supports Grok-3, Grok-3-mini, vision, and more.
Scanned 2/12/2026
Install via CLI
openskills install openclaw/skills---
name: xai
description: Chat with Grok models via xAI API. Supports Grok-3, Grok-3-mini, vision, and more.
homepage: https://docs.x.ai
triggers:
- grok
- xai
- ask grok
metadata:
clawdbot:
emoji: "🤖"
---
# xAI / Grok
Chat with xAI's Grok models. Supports text and vision.
## Setup
Set your API key in the skill config:
```bash
# Via clawdbot config
clawdbot config set skills.entries.xai.apiKey "xai-YOUR-KEY"
# Or environment variable
export XAI_API_KEY="xai-YOUR-KEY"
```
Get your API key at: https://console.x.ai
## Commands
### Chat with Grok
```bash
node {baseDir}/scripts/chat.js "What is the meaning of life?"
```
### Use a specific model
```bash
node {baseDir}/scripts/chat.js --model grok-3-mini "Quick question: 2+2?"
```
### Vision (analyze images)
```bash
node {baseDir}/scripts/chat.js --image /path/to/image.jpg "What's in this image?"
```
### 🔍 Search X/Twitter (Real-time!)
```bash
node {baseDir}/scripts/search-x.js "Remotion video framework"
node {baseDir}/scripts/search-x.js --days 7 "Claude AI tips"
node {baseDir}/scripts/search-x.js --handles @remotion_dev "updates"
```
Uses xAI Responses API with x_search tool for real X posts with citations.
### List available models
```bash
node {baseDir}/scripts/models.js
```
## Available Models
- `grok-3` - Most capable, best for complex tasks
- `grok-3-mini` - Fast and efficient
- `grok-3-fast` - Optimized for speed
- `grok-2-vision-1212` - Vision model for image understanding
## Example Usage
**User:** "Ask Grok what it thinks about AI safety"
**Action:** Run chat.js with the prompt
**User:** "Use Grok to analyze this image" (with attached image)
**Action:** Run chat.js with --image flag
**User:** "What Grok models are available?"
**Action:** Run models.js
## API Reference
xAI API Docs: https://docs.x.ai/api
## Environment Variables
- `XAI_API_KEY` - Your xAI API key (required)
- `XAI_MODEL` - Default model (optional, defaults to grok-3)
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
**Complete production-ready guide for Google Gemini embeddings API** This skill provides comprehensive coverage of the `gemini-embedding-001` model for generating text embeddings, including SDK usage, REST API patterns, batch processing, RAG integration with Cloudflare Vectorize, and advanced use cases like semantic search and document clustering. ---
Interview, source-challenge, verify, save, and ADR-gate fuzzy coding requests into Codex-ready implementation specs. Use when a feature, bugfix, refactor, migration, repo-wide change, or architecture task needs user-verified requirements, source-backed decisions, durable architecture decisions, acceptance criteria, validation commands, rollout notes, saved spec/ADR files, and a Codex execution prompt. Do not use when already fully specified or when the user wants direct implementation now.
Use when a repo needs CodeGraph plus ast-grep for Codex MCP setup, exploration, impact analysis, structural search, or safe refactor planning.