Junet.io

Setup

Configure Junet through the browser-based setup wizard

Accessing the Setup Wizard

After installing Junet, open your browser and navigate to:

http://your-server-ip

Or if you've configured a domain:

https://your-domain.com

You will be greeted with the setup wizard.

Step 1: Welcome Screen

Click Next to begin the setup process.

Step 2: LLM Configuration

Junet supports multiple Large Language Model (LLM) providers. Choose the provider that best fits your needs and budget.

Supported Providers

  • Azure OpenAI - Enterprise-grade OpenAI models through Microsoft Azure
  • OpenAI - Direct access to OpenAI's GPT models
  • Anthropic - Claude models from Anthropic
  • OpenRouter - Access to multiple LLM providers through a unified API
  • vLLM - Self-hosted open-source models
  • Ollama - Local LLM deployment for privacy-focused setups

You can change your LLM provider later in the Junet settings if needed.

Azure OpenAI Setup

Click to expand Azure OpenAI setup instructions

Azure OpenAI provides enterprise-grade access to OpenAI models with enhanced security and compliance features.

Required Information

  • API Key: Your Azure OpenAI API key
  • Endpoint: Your Azure OpenAI resource endpoint
  • Deployment Name: The name of your deployed model
  • API Version: The API version (e.g., 2024-02-15-preview)

How to Get the Required Data

  1. Create an Azure OpenAI Resource

    • Go to the Azure Portal
    • Search for "Azure OpenAI" in the search bar
    • Click Create and fill in the required information
    • Choose your subscription, resource group, and region
    • Wait for the deployment to complete

  2. Get Your API Key and Endpoint

    • Navigate to your Azure OpenAI resource
    • In the left sidebar, click on Keys and Endpoint
    • Copy one of the keys (KEY 1 or KEY 2)
    • Copy the Endpoint URL

  3. Deploy a Model

    • In your Azure OpenAI resource, go to Model deployments
    • Click Create new deployment
    • Select a model (e.g., gpt-4, gpt-35-turbo)
    • Give it a deployment name (you'll need this later)
    • Click Create

  4. Find Your API Version

    • The recommended API version is usually listed in the Azure OpenAI documentation
    • Common versions: 2024-02-15-preview, 2023-12-01-preview

Configuration Example

API Key: abc123def456ghi789jkl012mno345pq
Endpoint: https://your-resource.openai.azure.com/
Deployment Name: gpt-4-deployment
API Version: 2024-02-15-preview

OpenAI Setup

Click to expand OpenAI setup instructions

OpenAI provides direct access to GPT models including GPT-4, GPT-4 Turbo, and GPT-3.5 Turbo.

Required Information

  • API Key: Your OpenAI API key
  • Organization ID (optional): Your OpenAI organization ID

How to Get the Required Data

  1. Create an OpenAI Account

  2. Get Your API Key

    • Navigate to API Keys
    • Click Create new secret key
    • Give it a name (e.g., "Junet")
    • Copy the API key immediately (you won't be able to see it again)
    • Store it securely

  3. Get Your Organization ID (Optional)

  4. Add Credits

    • Make sure your account has sufficient credits
    • Go to Billing to add payment information

Configuration Example

API Key: sk-proj-abc123def456ghi789jkl012mno345pqr678stu901vwx
Organization ID: org-abc123def456 (optional)

Keep your API key secure and never share it publicly. Treat it like a password.

Anthropic Setup

Click to expand Anthropic setup instructions

Anthropic provides access to Claude models, known for their strong performance on complex reasoning tasks.

Required Information

  • API Key: Your Anthropic API key

How to Get the Required Data

  1. Create an Anthropic Account

  2. Get Your API Key

    • Navigate to API Keys
    • Click Create Key
    • Give it a name (e.g., "Junet")
    • Copy the API key immediately
    • Store it securely

  3. Add Credits

    • Go to Billing
    • Add payment information and purchase credits

Configuration Example

API Key: sk-ant-api03-abc123def456ghi789jkl012mno345pqr678stu901vwx

Claude models are particularly good at following instructions and handling long contexts (up to 200K tokens).

OpenRouter Setup

Click to expand OpenRouter setup instructions

OpenRouter provides unified access to multiple LLM providers through a single API, including OpenAI, Anthropic, Google, Meta, and more.

Required Information

  • API Key: Your OpenRouter API key

How to Get the Required Data

  1. Create an OpenRouter Account

  2. Get Your API Key

    • Navigate to Keys
    • Click Create Key
    • Give it a name (e.g., "Junet")
    • Copy the API key
    • Store it securely

  3. Add Credits

    • Go to Settings > Credits
    • Add payment information and purchase credits
    • Note: OpenRouter uses a pay-as-you-go model

  4. Choose Your Models

    • Browse available models at openrouter.ai/models
    • You can use any model available on OpenRouter with Junet

Configuration Example

API Key: sk-or-v1-abc123def456ghi789jkl012mno345pqr678stu901vwx

OpenRouter offers access to many models at competitive prices, making it a cost-effective option for experimentation.

vLLM Setup

Click to expand vLLM setup instructions

vLLM is a high-performance inference engine for self-hosted open-source LLMs, offering maximum privacy and control.

Required Information

  • API Endpoint: The URL where your vLLM server is running
  • API Key (optional): If you've configured authentication

How to Set Up vLLM

  1. Set Up a vLLM Server

    • Install vLLM on a server with GPU support
    • See the vLLM documentation for installation instructions

  2. Start the vLLM Server

    python -m vllm.entrypoints.openai.api_server \
  --model meta-llama/Llama-2-7b-chat-hf \
  --host 0.0.0.0 \
  --port 8000

  3. Get Your API Endpoint

    • Your endpoint will be: http://your-server-ip:8000
    • For local testing: http://localhost:8000

  4. Configure Authentication (Optional)

    • Add --api-key your-secret-key to the vLLM command
    • Use this API key in Junet's configuration

Configuration Example

API Endpoint: http://192.168.1.100:8000
API Key: your-secret-key (optional)

vLLM requires significant GPU resources. Make sure your server meets the hardware requirements for your chosen model.

Ollama Setup

Click to expand Ollama setup instructions

Ollama enables you to run open-source LLMs locally on your machine, perfect for privacy-focused deployments.

Required Information

  • API Endpoint: The URL where Ollama is running (usually http://localhost:11434)

How to Set Up Ollama

  1. Install Ollama

    • Visit ollama.ai
    • Download and install Ollama for your operating system
    • For Linux: 
      curl -fsSL https://ollama.ai/install.sh | sh

  2. Start Ollama

    • Ollama starts automatically after installation
    • To start manually: 
      ollama serve

  3. Pull a Model

    • Download a model to use: 
      ollama pull llama2
    • Popular models: llama2, mistral, codellama, phi

  4. Get Your API Endpoint

    • Default: http://localhost:11434
    • For remote access: http://your-server-ip:11434

Configuration Example

API Endpoint: http://localhost:11434

Ollama is completely free and runs locally, making it ideal for privacy-sensitive use cases. However, performance depends on your hardware.

Step 3: Embedding Model Configuration

Embedding models are used to convert text into vector representations for semantic search and retrieval tasks.

Choose Your Embedding Provider

Similar to LLM configuration, you can choose from multiple embedding model providers:

  • Azure OpenAI - Ada embeddings through Microsoft Azure
  • OpenAI - OpenAI's text-embedding models
  • Ollama - Local embedding models for privacy
  • Self-hosted - Custom embedding API endpoints

Embedding models are crucial for features like semantic search, document retrieval, and knowledge base functionality.

Azure OpenAI Embeddings

Click to expand Azure OpenAI embeddings setup

Use the same Azure OpenAI resource you configured earlier, but deploy an embedding model.

Required Information

  • API Key: Same as your LLM configuration
  • Endpoint: Same as your LLM configuration
  • Deployment Name: Name of your embedding model deployment
  • API Version: Same as your LLM configuration

How to Deploy an Embedding Model

  1. Go to your Azure OpenAI resource in the Azure Portal
  2. Navigate to Model deployments
  3. Click Create new deployment
  4. Select an embedding model (e.g., text-embedding-ada-002, text-embedding-3-small, text-embedding-3-large)
  5. Give it a deployment name (e.g., embeddings)
  6. Click Create

Configuration Example

Deployment Name: embeddings
Model: text-embedding-3-small

OpenAI Embeddings

Click to expand OpenAI embeddings setup

Use the same OpenAI API key you configured for LLMs.

Required Information

  • API Key: Same as your LLM configuration
  • Model: Choose an embedding model

Available Models

  • text-embedding-3-small - Cost-effective, good performance
  • text-embedding-3-large - Higher quality, more expensive
  • text-embedding-ada-002 - Legacy model, still reliable

Configuration Example

API Key: (use existing key)
Model: text-embedding-3-small

Ollama Embeddings

Click to expand Ollama embeddings setup

Use your existing Ollama installation with an embedding model.

Required Information

  • API Endpoint: Same as your Ollama LLM configuration
  • Model Name: The embedding model to use

Pull an Embedding Model

ollama pull nomic-embed-text

Popular embedding models:

  • nomic-embed-text - High quality, recommended
  • mxbai-embed-large - Large context window
  • all-minilm - Fast and lightweight

Configuration Example

API Endpoint: http://localhost:11434
Model: nomic-embed-text

Step 4: Single Sign-On (SSO) Configuration

Configure Single Sign-On to allow users to authenticate with their existing accounts, or skip this step to use local user accounts only.

You can skip this step and configure SSO later in the Junet settings if needed.

Supported SSO Providers

  • Google - Sign in with Google Workspace or Gmail accounts
  • Microsoft - Sign in with Microsoft 365 or Azure AD accounts

Google SSO Setup

Click to expand Google SSO setup instructions

Allow users to sign in with their Google accounts.

Required Information

  • Client ID: Your Google OAuth 2.0 Client ID
  • Client Secret: Your Google OAuth 2.0 Client Secret

How to Get the Required Data

  1. Go to Google Cloud Console

  2. Enable Google+ API

    • Go to APIs & Services > Library
    • Search for "Google+ API"
    • Click Enable

  3. Create OAuth Credentials

    • Go to APIs & Services > Credentials
    • Click Create Credentials > OAuth client ID
    • Select Web application as the application type
    • Give it a name (e.g., "Junet SSO")

  4. Configure Authorized Redirect URIs

    • Add your Junet callback URL: 
      https://your-junet-domain.com/api/auth/callback/google
    • For local testing: 
      http://localhost/api/auth/callback/google

  5. Get Client ID and Secret

    • After creation, copy the Client ID and Client Secret
    • Store them securely

Configuration Example

Client ID: 123456789-abc123def456.apps.googleusercontent.com
Client Secret: GOCSPX-abc123def456ghi789

Make sure to add the correct redirect URI, otherwise authentication will fail.

Microsoft SSO Setup

Click to expand Microsoft SSO setup instructions

Allow users to sign in with their Microsoft 365 or Azure AD accounts.

Required Information

  • Client ID (Application ID): Your Azure AD app registration ID
  • Client Secret: Your Azure AD app client secret
  • Tenant ID: Your Azure AD tenant ID (or use "common" for multi-tenant)

How to Get the Required Data

  1. Go to Azure Portal

    • Navigate to portal.azure.com
    • Go to Azure Active Directory (or Microsoft Entra ID)

  2. Register an Application

    • Click App registrations > New registration
    • Enter a name (e.g., "Junet SSO")
    • Select supported account types:
      • Single tenant: Only your organization
      • Multitenant: Any Azure AD directory
      • Multitenant + personal: Any Azure AD + personal Microsoft accounts
    • Click Register

  3. Configure Redirect URI

    • In your app registration, go to Authentication
    • Click Add a platform > Web
    • Add your Junet callback URL: 
      https://your-junet-domain.com/api/auth/callback/microsoft
    • For local testing: 
      http://localhost/api/auth/callback/microsoft
    • Click Configure

  4. Get Client ID and Tenant ID

    • Go to Overview in your app registration
    • Copy the Application (client) ID
    • Copy the Directory (tenant) ID

  5. Create a Client Secret

    • Go to Certificates & secrets
    • Click New client secret
    • Add a description (e.g., "Junet Secret")
    • Select an expiration period
    • Click Add
    • Copy the secret Value immediately (you won't see it again)

  6. Configure API Permissions (Optional)

    • Go to API permissions
    • Add Microsoft Graph > Delegated permissions
    • Add: User.Read, email, profile, openid
    • Click Grant admin consent if required

Configuration Example

Client ID: 12345678-1234-1234-1234-123456789abc
Client Secret: abc~123.def_456-ghi789jkl012
Tenant ID: 87654321-4321-4321-4321-cba987654321
(or "common" for multi-tenant)

Using "common" as Tenant ID allows users from any Azure AD organization to sign in. Use your specific Tenant ID to restrict access to your organization only.

Skip SSO Configuration

If you prefer to use only local user accounts without SSO integration, simply click Skip to continue to the next step.

You can always configure SSO later in Settings > Authentication.

Step 5: Primary URL Configuration

Set the primary URL where Junet will be accessible. This is important for generating correct links and callbacks.

Enter Your Primary URL

Provide the URL where your users will access Junet:

Examples:

  • https://chat.company.com
  • https://junet.mycompany.io
  • https://ai.example.com

Make sure this URL matches your actual domain configuration and SSL/HTTPS setup. Incorrect URLs can cause authentication and callback issues.

Configuration Example

Primary URL: https://chat.company.com

Step 6: First Agent Configuration (Optional)

Agents are AI assistants that can be configured with specific instructions, tools, and knowledge bases to handle specialized tasks.

Create Your First Agent

You can create an agent now or skip this step and create agents later from the Junet dashboard.

Select an Agent Type

Choose from pre-configured agent templates or create a custom agent:

Available Agent Templates:

  • Jira - Connect to Jira for issue tracking and project management
  • Confluence - Access Confluence spaces for documentation retrieval
  • MediaWiki - Connect to MediaWiki installations for knowledge base access
  • Generic API - Connect to any custom REST API endpoint
  • Custom Agent - Build your own agent from scratch

Agent templates come with pre-configured tools and capabilities specific to each service.

Agent Configuration

After selecting an agent type, configure the following:

Basic Information:

  • Name: Give your agent a descriptive name that reflects its purpose
    • Example: Confluence DevOps Space
    • Example: Jira Project Management
    • Example: Internal Wiki Assistant
  • Description: Detailed description used by the router for decision-making
    • Example: Confluence DevOps Space for CI/CD documentation and deployment guides
    • Example: Jira agent for managing DevOps team tickets and sprint planning
    • This helps Junet's routing system decide which agent to use for specific queries

Connection Details (varies by agent type):

Jira Agent Configuration

Required Connection Details

  • Jira URL: Your Jira instance URL (e.g., https://company.atlassian.net)
  • Username/Email: Your Jira account email
  • API Token: Jira API token for authentication
  • Project Key (optional): Specific project to focus on (e.g., DEV, OPS)

How to Get a Jira API Token

  1. Go to Atlassian Account Settings
  2. Click Create API token
  3. Give it a label (e.g., "Junet Integration")
  4. Copy the token immediately and store it securely

Configuration Example

Name: Jira DevOps Tickets
Description: Jira agent for DevOps team issue tracking, sprint management, and bug reports
Jira URL: https://company.atlassian.net
Username: devops@company.com
API Token: ATATT3xFfGF0abcdef123456789
Project Key: DEVOPS
Confluence Agent Configuration

Required Connection Details

  • Confluence URL: Your Confluence instance URL (e.g., https://company.atlassian.net/wiki)
  • Username/Email: Your Confluence account email
  • API Token: Confluence API token for authentication
  • Space Key (optional): Specific space to focus on (e.g., DEV, DOCS)

How to Get a Confluence API Token

  1. Go to Atlassian Account Settings
  2. Click Create API token
  3. Give it a label (e.g., "Junet Confluence")
  4. Copy the token immediately and store it securely

Configuration Example

Name: Confluence DevOps Space
Description: Confluence DevOps Space for CI/CD documentation, deployment procedures, and infrastructure guides
Confluence URL: https://company.atlassian.net/wiki
Username: devops@company.com
API Token: ATATT3xFfGF0abcdef123456789
Space Key: DEVOPS
MediaWiki Agent Configuration

Required Connection Details

  • Wiki URL: Your MediaWiki instance URL (e.g., https://wiki.company.com)
  • API Endpoint: MediaWiki API path (usually /api.php)
  • Username: MediaWiki username
  • Password or Bot Token: Authentication credentials
  • Namespace (optional): Specific namespace to focus on

Configuration Example

Name: Internal Wiki Documentation
Description: Company internal wiki for technical documentation, how-to guides, and knowledge base articles
Wiki URL: https://wiki.company.com
API Endpoint: /api.php
Username: junet-bot
Bot Token: abc123def456ghi789
Namespace: Technical_Documentation
Generic API Agent Configuration

Required Connection Details

  • API Base URL: Base URL of your API (e.g., https://api.company.com)
  • Authentication Method: Choose from API Key, Bearer Token, Basic Auth, or OAuth
  • API Key/Token: Authentication credentials
  • Custom Headers (optional): Additional HTTP headers required by your API
  • Endpoints: Define available API endpoints and their purposes

Configuration Example

Name: Internal CRM API
Description: Custom CRM system for customer data, support tickets, and sales information
API Base URL: https://api.company.com/v1
Authentication: Bearer Token
Token: bearer_abc123def456ghi789jkl012
Custom Headers: X-Company-ID: 12345
Available Endpoints:
  - /customers - Customer information
  - /tickets - Support tickets
  - /sales - Sales data

Advanced Settings:

  • System Prompt: Custom instructions that define the agent's behavior and personality
  • Model: Select which LLM model this agent should use
  • Temperature: Control randomness in responses (0.0 = focused, 1.0 = creative)
  • Tools: Enable additional tools like web search, code execution, or file access
  • Knowledge Base: Attach additional documents or data sources

Important: The Name and Description are used by Junet's intelligent routing system to automatically direct user queries to the most appropriate agent. Be specific and descriptive!

Don't worry if you're not ready to configure an agent yet. You can create and customize agents anytime from the Junet dashboard.

Skip Agent Creation

Click Skip if you want to set up agents later. You'll be able to create multiple agents with different configurations after setup is complete.

Setup Complete! 🎉

Congratulations! Junet is now ready to use.

Click Done to access your Junet dashboard and start using your AI platform.

What's Next?

After completing the setup, you can:

  • Create Agents: Build specialized AI assistants for different tasks
  • Invite Users: Add team members and manage permissions
  • Upload Knowledge: Add documents and data sources for your agents
  • Integrate Tools: Connect external services and APIs
  • Customize Settings: Fine-tune configurations and preferences

For more information, check out the Configuration Documentation.

Related Articles

Installation
Requirements
Configure an Agent
Previous
Installation
Next
Agents & Connections
Setup | Junet.io Documentation