How to Connect Cadreen as a Model Provider

Cadreen speaks the OpenAI API format. Any tool that works with any OpenAI-compatible provider works with Cadreen — just change the base URL. You get governance, memory, traces, and self-healing on top.

Quick Start

Base URLhttps://accomplishanything.today/api/v1/cadreen

API Keysk_cadreen_...— get from Settings

Model"cadreen"— required by clients, Cadreen ignores it

curl

BASH

curl -X POST https://accomplishanything.today/api/v1/cadreen/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk_cadreen_..." \
  -d '{
    "model": "cadreen",
    "messages": [
      {"role": "user", "content": "Hello"}
    ]
  }'}

Python

TEXT

from openai import OpenAI

client = OpenAI(
    base_url="https://accomplishanything.today/api/v1/cadreen",
    api_key="sk_cadreen_...",
)

response = client.chat.completions.create(
    model="cadreen",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Node.js

TEXT

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://accomplishanything.today/api/v1/cadreen",
  apiKey: "sk_cadreen_...",
});

const response = await client.chat.completions.create({
  model: "cadreen",
  messages: [{ role: "user", content: "Hello" }],
});
console.log(response.choices[0].message.content);

Go

TEXT

client := openai.NewClient(
    "sk_cadreen_...",
    "https://accomplishanything.today/api/v1/cadreen",
)

resp, err := client.Chat.Completions.New(ctx, openai.ChatCompletionNewParams{
    Model: "cadreen",
    Messages: []openai.ChatCompletionMessageParamUnion{
        openai.UserMessage("Hello"),
    },
})

Note

The model field is required by OpenAI-compatible clients but Cadreen ignores it. Cadreen decides which model to use internally based on the request.

Platform-Specific Setup

Cursor

TEXT

// Cursor Settings → Models → Add Model
{
  "name": "Cadreen",
  "baseURL": "https://accomplishanything.today/api/v1/cadreen",
  "model": "cadreen",
  "apiKey": "sk_cadreen_..."
}

opencode

TEXT

// opencode.json
{
  "provider": {
    "name": "@ai-sdk/openai-compatible",
    "baseURL": "https://accomplishanything.today/api/v1/cadreen/v1",
    "apiKey": "sk_cadreen_..."
  },
  "model": "cadreen/cadreen"
}

Continue

TEXT

# config.yaml
models:
  - name: Cadreen
    provider: openai
    baseURL: https://accomplishanything.today/api/v1/cadreen
    model: cadreen
    apiKey: sk_cadreen_...

LangChain

TEXT

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="cadreen",
    openai_api_base="https://accomplishanything.today/api/v1/cadreen",
    openai_api_key="sk_cadreen_...",
)

LlamaIndex

TEXT

from llama_index.llms.openai import OpenAI

llm = OpenAI(
    model="cadreen",
    api_base="https://accomplishanything.today/api/v1/cadreen",
    api_key="sk_cadreen_...",
)

LiteLLM

TEXT

import litellm

response = litellm.completion(
    model="openai/cadreen",
    api_base="https://accomplishanything.today/api/v1/cadreen",
    api_key="sk_cadreen_...",
    messages=[{"role": "user", "content": "Hello"}],
)

Vercel AI SDK

TEXT

import { createOpenAI } from "@ai-sdk/openai";

const cadreen = createOpenAI({
  baseURL: "https://accomplishanything.today/api/v1/cadreen",
  apiKey: "sk_cadreen_...",
});

const model = cadreen("cadreen");

What Cadreen Adds

You don't choose Cadreen instead of your model. You choose structure around it. Same API, same tools — but every request is governed, remembered, and traced.

GovernanceEvery tool call checked against your rules. Approvals, blocks, escalations — you decide.

MemoryRemembers across sessions. Facts, preferences, past decisions — searchable and versioned.

Intelligence TracesSee why every decision was made. What it checked, what it remembered, what it decided.

Self-HealingRetries failed tool calls automatically. Analyzes errors, fixes arguments, tries again.

Auto-LearningTurns successful patterns into permanent knowledge. Gets better over time.

What Cadreen Adds Over Raw API Calls

temperature, top_pIgnored — Cadreen manages internally

model fieldRequired but ignored — use "cadreen"

Response metadataIncludes confidence, governance decision, trace ID

Chain-of-thoughtStripped — not visible to client

n parameterAlways 1 — one completion per request

Next steps:

Set up governance rules Tool calling deep dive