Quickstart

Three steps, then your first request from whichever tool you use.

Create an account and, under Dashboard → Providers, add a model key (Anthropic, OpenAI, Google, an OpenAI-compatible endpoint, or a Claude Code token). An OpenAI or Google key also provides embeddings for memory.
Create an API key under Dashboard → API keys. It starts with sk-memtro- and is shown once.
Send a request:

curl https://platform.memtro.com/v1/chat/completions \
  -H "Authorization: Bearer $MEMTRO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "Who owns the billing migration?"}],
    "memtro": {"mode": "agent"}
  }'

from openai import OpenAI

client = OpenAI(base_url="https://platform.memtro.com/v1", api_key="sk-memtro-...")

stream = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Have a look at Help Scout ticket #2450"}],
    stream=True,
    extra_body={"memtro": {"mode": "agent"}},
)
for chunk in stream:
    delta = chunk.choices[0].delta if chunk.choices else None
    if delta and delta.content:
        print(delta.content, end="")

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://platform.memtro.com/v1", apiKey: process.env.MEMTRO_API_KEY });

const stream = await client.chat.completions.create({
  model: "auto:agent",
  messages: [{ role: "user", content: "Summarise open VIP tickets older than a day" }],
  stream: true,
});
for await (const chunk of stream) process.stdout.write(chunk.choices[0]?.delta?.content ?? "");

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "enabled": true,
      "headers": { "Authorization": "Bearer sk-memtro-..." }
    }
  }
}

Then, in OpenCode: "Use memtro to find who owns the billing migration."

claude mcp add --transport http memtro https://platform.memtro.com/api/mcp \
  --header "Authorization: Bearer sk-memtro-..."

Then, in Claude Code: "Search memtro for what we decided about the Q3 launch."

What happens next

auto routes the request by difficulty; memtro uses your organisation's default model. See Model routing.
"memtro": {"mode": "agent"} (or :agent on the model id) runs a multi-step investigation across your connected systems. See Agent mode.
Connect your systems on Dashboard → Connections so the model has something to look up. See Connectors.
Facts from the exchange are extracted into memory, personal or shared. See Memory.