平台开发与集成指南
概览
Base URL
https://api.aizhs.top/v1 # 生产
https://api-staging.aizhs.top/v1 # 预发
http://localhost:8802/v1 # 自托管认证方式
Authorization: Bearer <jwt>X-API-Key: <key>,在 /settings/api-keys 生成X-Webhook-Signature: <hmac>速率限制
默认 60 req/min,Pro 200/min,Team 500/min。超限返回 429,响应头含:
X-RateLimit-Limit: 60
X-RateLimit-Remaining: 0
X-RateLimit-Reset: 1699900000OpenAI 兼容接口,支持流式(SSE)和非流式。
curl -X POST https://api.aizhs.top/v1/chat/completions \
-H "Authorization: Bearer <jwt>" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o",
"messages": [
{"role": "user", "content": "你好"}
],
"stream": true,
"temperature": 0.7
}'
# 流式响应(SSE)
data: {"id":"...","choices":[{"delta":{"content":"你"},"index":0}]}
data: {"id":"...","choices":[{"delta":{"content":"好"},"index":0}]}
data: [DONE]调用指定 Agent(带知识库 + MCP 工具 + 工作流)。
curl -X POST https://api.aizhs.top/v1/chat/agents/agent_abc123 \
-H "X-API-Key: <key>" \
-H "Content-Type: application/json" \
-d '{
"message": "帮我分析这份简历",
"files": ["file_xyz"],
"context": { "userId": "u_123" }
}'统一响应格式:
{
"code": 0, // 0=成功, 非 0=失败
"message": "ok",
"data": { ... } // 成功时的数据
}npm i @ihui/api-clientimport { chat } from '@ihui/api-client'
await chat({ message: '你好' })pip install ihui-aifrom ihui import IHUI
client = IHUI(api_key='...')
client.chat(message='你好')go get github.com/ihui/ihui-goclient := ihui.New("key")
client.Chat(ctx, "你好")