# apifare tools for OpenAI GPT

OpenAI GPT is a model, not an app, so it gets apifare through whatever runs it. Get a token first: https://apifare.com/start?origin=page:models-gpt

1. Run OpenAI GPT in an MCP client: OpenAI Codex CLI (/integrations/codex) runs OpenAI models and speaks MCP, and so do Cursor (/integrations/cursor) and GitHub Copilot (/integrations/vscode).
2. Or hand it apifare as tools over HTTP (OpenAI's OpenAI-compatible API):

```python
import json, os, requests
from openai import OpenAI

llm = OpenAI(
    api_key=os.environ["OPENAI_API_KEY"],
)
APIFARE = "https://apifare.com"
auth = {"authorization": "Bearer " + os.environ["APIPAY_TOKEN"]}

def tool(name, description, properties):
    return {"type": "function", "function": {
        "name": name, "description": description,
        "parameters": {"type": "object", "properties": properties,
                       "required": list(properties)}}}

tools = [
    tool("apifare_search", "Find a paid API capability by what it does",
         {"q": {"type": "string"}}),
    tool("apifare_call", "Call a capability by slug with its params",
         {"slug": {"type": "string"}, "params": {"type": "object"}}),
]

def run(name, args):
    if name == "apifare_search":
        r = requests.get(APIFARE + "/v1/search",
                         params={"q": args["q"]}, headers=auth)
    else:
        r = requests.post(APIFARE + "/v1/call/" + args["slug"],
                          json=args["params"], headers=auth)
    return r.json()

messages = [{"role": "user",
             "content": "Search the web for today's top AI news"}]
while True:
    msg = llm.chat.completions.create(
        model="<model>", messages=messages, tools=tools,
    ).choices[0].message
    if not msg.tool_calls:
        print(msg.content)
        break
    messages.append(msg)
    for call in msg.tool_calls:
        result = run(call.function.name, json.loads(call.function.arguments))
        messages.append({"role": "tool", "tool_call_id": call.id,
                         "content": json.dumps(result)})
```

Set OPENAI_API_KEY and APIPAY_TOKEN, and use a tool-calling model from https://developers.openai.com/api/docs/models.
3. Caps and the kill switch are checked before dispatch; an empty balance returns HTTP 402 with a top-up link (see /errors.md).

Endpoint and tool-calling format checked against https://developers.openai.com/api/docs/guides/function-calling on 2026-09-12.