Quickstart

Create a BlueNexus account

Go to app.bluenexus.ai and sign up. New accounts start with 100,000 credits ($10.00).

Connect a service

In the dashboard, go to Connections and connect at least one service (e.g., Google Workspace, Slack, or GitHub). This is the data your AI agent will access.

Create a Personal Access Token

Go to Settings > Sessions and click Create Personal Access Token. Select the universal-mcp-read-write scope. Save the token — you'll only see it once.

Configure your MCP client

Point your MCP client at the BlueNexus endpoint:

{
  "mcpServers": {
    "bluenexus": {
      "type": "streamable-http",
      "url": "https://api.bluenexus.ai/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_PERSONAL_ACCESS_TOKEN"
      }
    }
  }
}

Settings > Apps > Advanced Settings > Enable Developer Mode > Create App

Name: BlueNexus
MCP Server URL: https://api.bluenexus.ai/mcp
Authentication: OAuth
Leave Client ID and Client Secret empty
Check "I understand and want to continue"

Server URL: https://api.bluenexus.ai/mcp
Transport: Streamable HTTP
Auth Header: Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN

curl -X POST https://api.bluenexus.ai/mcp \
  -H "Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -H "X-Response-Format: json" \
  -d '{"jsonrpc":"2.0","method":"tools/list","params":{},"id":"1"}'

Make your first tool call

Ask your AI assistant something that uses your connected service:

Behind the scenes, the MCP client sends:

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "name": "use-agent",
    "arguments": {
      "prompt": "What's on my Google Calendar today?",
      "connector": "google"
    }
  },
  "id": "1"
}

The response contains the agent's answer with data from your connected service.

Create a BlueNexus account

Go to app.bluenexus.ai and sign up.

Create a Personal Access Token

Go to Settings > Sessions and click Create Personal Access Token. Select the scopes you need:

• llm-all — LLM chat completions
• agents-use — Use AI agents
• connections — Manage connections
• universal-mcp-read-write — MCP access

Call the Chat Completions API

BlueNexus offers an OpenAI-compatible chat completions endpoint:

curl -X POST https://api.bluenexus.ai/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bluenexus/glm-4.7-flash-tee",
    "messages": [
      {"role": "user", "content": "Hello, what models are available?"}
    ]
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_PERSONAL_ACCESS_TOKEN",
    base_url="https://api.bluenexus.ai/api/v1"
)

response = client.chat.completions.create(
    model="bluenexus/glm-4.7-flash-tee",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_PERSONAL_ACCESS_TOKEN",
  baseURL: "https://api.bluenexus.ai/api/v1",
});

const response = await client.chat.completions.create({
  model: "bluenexus/glm-4.7-flash-tee",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

The response includes credit headers: X-Credits-Consumed and X-Credits-Remaining.

List available models

curl https://api.bluenexus.ai/api/v1/models \
  -H "Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN"

Chat with an agent

If you have an agent configured (or use the default), chat with it:

curl -X POST https://api.bluenexus.ai/api/v1/agents/default/chat/completions \
  -H "Authorization: Bearer YOUR_PERSONAL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {"role": "user", "content": "Summarize my unread Slack messages"}
    ],
    "stream": false
  }'

The default agent ID resolves to your account's default agent. You can also create agents via the Agent Builder UI or the API.