skip to content
conifer
Overview

CONIFER · DOCS

SDK

Call the gateway

Two supported clients: ours, for cost control, or the official OpenAI and Anthropic packages, for a drop-in.

Every path below uses one credential and one host. Mint a key at the console, store it as CONIFER_API_KEY, and point the client at https://api.conifer.build.

terminal
export CONIFER_API_KEY='sk-conifer-…'

Which client

ClientUse it when
The Conifer SDKYou want the exact cost of each turn, a hard spend ceiling, or errors you can branch on. Open source.The Conifer SDK
Official OpenAI SDKAn existing codebase, or a plain drop-in. Chat completions, Responses, streaming, tools.TypeScript
Official Anthropic SDKThe Messages wire, Anthropic model ids, prompt cache breakpoints.Python
Environment drop-inCode you do not want to touch at all. Set two variables.Replace your provider key
curlA one-off, a script, or a language with no SDK yet.API reference
MCPAn agent, a bot, or an editor that speaks MCP rather than the OpenAI wire.MCP and plugins

The Conifer SDK is open source

It lives at ConiferKit/use-conifer in TypeScript and Python, and it exists for the things the OpenAI client structurally cannot do: an exact integer cost receipt on every response, a server-enforced maxCostNanoUsd ceiling, and distinct error types for “out of credit” versus “your own ceiling refused this”. If an integration is wrong, you can read why and send a fix — Contributing.

Or the official packages

The gateway speaks the OpenAI and Anthropic wires, so the official clients already know how to talk to it. Nothing about the request body changes; only the base URL and the key do.

terminal
npm install openai
route.ts
import OpenAI from "openai";

export function conifer() {
  const apiKey = process.env.CONIFER_API_KEY;
  if (!apiKey) {
    throw new Error("CONIFER_API_KEY is missing. Set it before calling the gateway.");
  }
  return new OpenAI({
    baseURL: process.env.OPENAI_BASE_URL ?? "https://api.conifer.build/v1",
    apiKey,
  });
}

export async function POST(req: Request) {
  const res = await conifer().chat.completions.create({
    model: "claude-haiku-4-5",
    max_tokens: 1024,
    messages: [{ role: "user", content: "three names for a build cache" }],
  });

  return Response.json({ content: res.choices[0].message.content });
}

If the project already builds its client from environment variables, set these and leave the source alone. Note the asymmetry: the OpenAI base URL carries the /v1 suffix and the Anthropic one does not.

terminal
export OPENAI_BASE_URL=https://api.conifer.build/v1
export OPENAI_API_KEY=$CONIFER_API_KEY
terminal
export ANTHROPIC_BASE_URL=https://api.conifer.build
export ANTHROPIC_API_KEY=$CONIFER_API_KEY