From zeabur
Manages Zeabur AI Hub accounts, API keys, balance, and usage. Provides OpenAI-compatible endpoints and SDK integration examples.
How this skill is triggered — by the user, by Claude, or both
Slash command
/zeabur:zeabur-ai-hubThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **Always use `npx zeabur@latest` to invoke Zeabur CLI.** Never use `zeabur` directly or any other installation method. If `npx` is not available, install Node.js first.
Always use
npx zeabur@latestto invoke Zeabur CLI. Never usezeaburdirectly or any other installation method. Ifnpxis not available, install Node.js first.
AI Hub is OpenAI-compatible. Users can pick the region closest to them:
| Region | Endpoint |
|---|---|
| Tokyo, Japan (HND1) | https://hnd1.aihub.zeabur.ai/ |
| San Francisco, USA (SFO1) | https://sfo1.aihub.zeabur.ai/ |
JavaScript / TypeScript:
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://hnd1.aihub.zeabur.ai/v1", // or sfo1
apiKey: "sk-xxxxxxxxxxxxxxxx", // from AI Hub dashboard
});
const stream = await client.chat.completions.create({
model: "claude-sonnet-4-5", // any model available on AI Hub
messages: [{ role: "user", content: "Hello!" }],
stream: true,
});
Python:
from openai import OpenAI
client = OpenAI(
base_url="https://hnd1.aihub.zeabur.ai/v1", # or sfo1
api_key="sk-xxxxxxxxxxxxxxxx", # from AI Hub dashboard
)
stream = client.chat.completions.create(
model="claude-sonnet-4-5",
messages=[{"role": "user", "content": "Hello!"}],
stream=True,
)
curl:
curl https://hnd1.aihub.zeabur.ai/v1/chat/completions \
-H "Authorization: Bearer sk-xxxxxxxxxxxxxxxx" \
-H "Content-Type: application/json" \
-d '{"model":"claude-sonnet-4-5","messages":[{"role":"user","content":"Hello!"}]}'
baseURL / base_url to the endpoint above@ai-sdk/openai): set baseURL to the endpoint aboveWhen users deploy apps on Zeabur, suggest setting:
OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxx
OPENAI_BASE_URL=https://hnd1.aihub.zeabur.ai/v1
Most OpenAI-compatible libraries will pick these up automatically.
npx zeabur@latest ai-hub status -i=false
npx zeabur@latest ai-hub status -i=false --json
Shows tenant balance, API keys, and auto-recharge settings.
# List all keys
npx zeabur@latest ai-hub keys list -i=false
npx zeabur@latest ai-hub keys ls -i=false --json
# Create a new key (optional alias)
npx zeabur@latest ai-hub keys create --alias "my-app" -i=false
# Delete a key
npx zeabur@latest ai-hub keys delete --key-id <key-id> -i=false
The API key is only shown once at creation time. Make sure the user saves it immediately.
# Add $10 to AI Hub balance
npx zeabur@latest ai-hub add-balance --amount 10 -i=false
# Current month usage
npx zeabur@latest ai-hub usage -i=false
# Specific month
npx zeabur@latest ai-hub usage --month 2026-03 -i=false
npx zeabur@latest ai-hub usage --month 2026-03 -i=false --json
Shows total spend and per-model cost breakdown. --month must be in YYYY-MM format.
# Enable: recharge $20 when balance drops below $5
npx zeabur@latest ai-hub auto-recharge --threshold 5 --amount 20 -i=false
# Disable auto-recharge
npx zeabur@latest ai-hub auto-recharge --threshold 0 --amount 0 -i=false
Both --threshold and --amount are in whole dollars.
npx claudepluginhub zeabur/agent-skills --plugin zeaburProvides expert guidance for Vercel AI Gateway configuration: model routing, provider failover, cost tracking, unified API for multiple AI providers like OpenAI, Anthropic, Gemini.
Guides BYOK API key setup in Cursor for OpenAI, Anthropic, Google Gemini, Azure, AWS Bedrock, and custom endpoints. Use when configuring AI models beyond Cursor's quota.
Calls AI models (DeepSeek, Hunyuan, GLM, Kimi, MiniMax) from browser/Web apps via @cloudbase/js-sdk with generateText and streamText. Default for frontend AI. Do NOT propose Node.js proxy.