Ainative Chat Completions

Help agents build conversational AI with AINative's Chat Completions API. Use when (1) Building a chatbot or AI assistant, (2) Implementing streaming respons...

Security Scan

This is an instruction-only SDK / API documentation for AINative Chat Completions; its examples and requirements are consistent with that purpose and it does not request extra credentials or install anything unexpected.

Purpose & Capability

The name and description match the SKILL.md content: examples for raw HTTP, Python, React and Next.js SDK usage of AINative's chat completions API. The skill does not declare unrelated env vars, binaries, or installs, which is proportionate for documentation/example content.

Instruction Scope

Instructions are limited to making HTTP requests, using SDK hooks, and streaming SSE; they do not instruct reading arbitrary files or unrelated system state. One caveat: the examples include system messages (e.g., "You are a helpful assistant."), which is expected for chat-completion examples but triggered a pattern scanner for 'system-prompt-override' — likely a benign example but worth being aware of (see scan findings).

Install Mechanism

No install spec and no bundled code — instruction-only. This minimizes risk because nothing is written to disk by the skill itself.

Credentials

The SKILL.md shows use of an AINative API key (X-API-Key or process.env.AINATIVE_API_KEY) in examples. The skill does not require or request unrelated credentials or system config paths; requiring an API key is proportional to the documented usage.

Persistence & Privilege

always is false and there is no install step or code that requests persistent presence or modifies other skills or system settings.

Scan Findings in Context

[system-prompt-override] expected: The SKILL.md contains example system messages (e.g., "You are a helpful assistant.") and frontmatter metadata; that pattern can trigger a 'system-prompt-override' detector even though here it's used as a normal example. Treat this as a likely false positive but remain cautious about any hidden or elevated system-prompt instructions in third-party skills.

Assessment

This skill is documentation/example code for AINative's chat completions and is internally consistent. Before using it: (1) be prepared to supply your AINative API key when you run the examples — the skill itself doesn't store or request keys but the API calls require them; (2) do not paste secrets or sensitive data into prompts that will be sent to the AINative API; (3) verify you install SDK packages (@ainative/react-sdk, @ainative/next-sdk) from trusted sources (npmjs.org or your vetted private registry) and confirm the API endpoint (api.ainative.studio) matches the official provider; (4) note the scanner flagged a 'system-prompt-override' pattern — this appears to be a benign example of a system message, but always review any skill content that instructs agents on system prompts or message manipulation; (5) if you need higher assurance, request the skill author/publisher identity and confirm the SDK package repositories and docs referenced in the SKILL.md are official.

Endpoint

POST https://api.ainative.studio/v1/public/chat/completions

Auth: X-API-Key: ak_... or Authorization: Bearer <jwt>

Basic Request

import requests

response = requests.post(
    "https://api.ainative.studio/v1/public/chat/completions",
    headers={"X-API-Key": "ak_your_key"},
    json={
        "model": "claude-3-5-sonnet-20241022",
        "messages": [
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": "What is ZeroDB?"}
        ],
        "temperature": 0.7,
        "max_tokens": 1024
    }
)
data = response.json()
print(data["choices"][0]["message"]["content"])
print(f"Tokens used: {data['usage']['total_tokens']}")

Streaming (SSE)

import requests

with requests.post(
    "https://api.ainative.studio/v1/public/chat/completions",
    headers={"X-API-Key": "ak_your_key"},
    json={"model": "claude-3-5-sonnet-20241022",
          "messages": [{"role": "user", "content": "Count to 5"}],
          "stream": True},
    stream=True
) as resp:
    for line in resp.iter_lines():
        if line and line.startswith(b"data: "):
            chunk = line[6:]
            if chunk != b"[DONE]":
                import json
                delta = json.loads(chunk)["choices"][0]["delta"]
                print(delta.get("content", ""), end="", flush=True)

React SDK — useChat Hook

npm install @ainative/react-sdk

import { AINativeProvider, useChat } from '@ainative/react-sdk';

function App() {
  return (
    <AINativeProvider config={{ apiKey: 'ak_your_key' }}>
      <ChatComponent />
    </AINativeProvider>
  );
}

function ChatComponent() {
  const { messages, sendMessage, isLoading, error } = useChat({
    model: 'claude-3-5-sonnet-20241022',
    systemPrompt: 'You are a helpful assistant.',
  });

  const handleSend = () => sendMessage('Hello!');

  return (
    <div>
      {messages.map((msg, i) => (
        <div key={i}><b>{msg.role}:</b> {msg.content}</div>
      ))}
      <button onClick={handleSend} disabled={isLoading}>
        {isLoading ? 'Thinking...' : 'Send'}
      </button>
      {error && <p>Error: {error.message}</p>}
    </div>
  );
}

Next.js — Server Action + Streaming

npm install @ainative/next-sdk

// app/api/chat/route.ts
import { createServerClient } from '@ainative/next-sdk/server';

export async function POST(request: Request) {
  const { messages } = await request.json();
  const client = createServerClient({ apiKey: process.env.AINATIVE_API_KEY! });

  const stream = await client.chat.completions.create({
    model: 'claude-3-5-sonnet-20241022',
    messages,
    stream: true,
  });

  return new Response(stream.body, {
    headers: { 'Content-Type': 'text/event-stream' }
  });
}

Available Models

Model	Context	Best For
claude-3-5-sonnet-20241022	200k	General purpose (default)
claude-3-5-haiku-20241022	200k	Fast, low cost
claude-3-opus-20240229	200k	Complex reasoning
gpt-4o	128k	OpenAI fallback

Request Parameters

Parameter	Type	Default	Description
model	string	required	Model ID
messages	array	required	Conversation history
stream	boolean	false	Enable SSE streaming
temperature	float	0.7	Randomness (0-1)
max_tokens	int	1024	Max response length
system	string	—	System prompt (alternative)

Credit Costs

Credits are consumed per token. Check balance before high-volume usage:

balance = requests.get(
    "https://api.ainative.studio/api/v1/public/credits/balance",
    headers={"X-API-Key": "ak_your_key"}
).json()
print(f"Remaining: {balance['remaining_credits']}")

Error Handling

Status	Meaning	Action
401	Invalid API key	Check key format: ak_...
402	Insufficient credits	Top up or upgrade plan
429	Rate limit	Back off, retry with exponential delay
503	Model unavailable	Retry or use fallback model

References

• docs/api/CHAT_COMPLETION_API_REFERENCE.md (757 lines — full spec)
• src/backend/app/api/v1/endpoints/managed_chat.py
• packages/sdks/react/src/hooks/useChat.ts
• packages/sdks/nextjs/src/