# Welcome Leading AI Safety, Security, and Compliance

**Integrating High-Risk AI Securely.** Artificial Intelligence Risk, Inc. accelerates high-risk AI adoption safely by providing AI specifically for finance and other regulated industries, leveraging the best AI models, and wrapping it in our governance, risk, compliance, and cybersecurity layer. # Getting Started Step-by-step guide to setting up the AI Risk platform in your company. # Inviting Users Step-by-step guide on adding users to your company To invite new users, you must be a company admin. * [ ] You must have enough licenses purchased to accommodate the new user To invite a user, follow these steps: 1. Visit . 2. Click the 'Admin Console.' 3. Click the 'Admin' button. 4. Click the 'Manage Users' button. 5. Click 'Invite User.' 6. Fill out the 'Invite User' form. 7. Select a default agent for the user. 8. Click the 'Save' button.

7. Once the invite has been sent you will see the new user in the users list. Click the newly added user and assign them a role within the company. For more information on roles click [here](/admin/users/roles). # Adding New Large Language Models Over the next year, we plan to expand our LLM APIs by introducing about one new Endpoint every month. At present, our support extends to Azure OpenAI, OpenAI, and Mistral. 1. [**Azure Foundry**:](/getting-started/adding-new-large-language-models/azure-foundry) Microsoft’s end-to-end platform for building, customizing, and deploying AI applications on Azure. It brings together foundation models, data integration, and AI development tools in a single, secure environment. Teams can evaluate models, ground them in their own data, and deploy AI systems with built-in safety and governance. Azure AI Foundry is designed to help organizations move AI from prototype to production with confidence.. 2. [**OpenAI**](/getting-started/adding-new-large-language-models/openai): An independent research organization. OpenAI aims to ensure that artificial general intelligence benefits all of humanity. It develops and applies a suite of AI technologies, and its models have wide-ranging uses, from drafting emails or pieces of writing, to coding assistance, to answering questions. 3. [**Mistral AI**](/getting-started/adding-new-large-language-models/mistral): A powerful and fast model adaptable to many use-cases. While being 6x faster, it matches or outperform Llama 2 70B on all benchmarks, speaks many languages, has natural coding abilities. It handles 32k sequence length. You can use it through our API 4. LLama: (Large Language Model Meta AI, formerly stylized as LLaMA) is a family of autoregressive large language models (LLMs) released by Meta AI starting in February 2023. The latest version is Llama 3.3, released in December 2024. 5. Amazon Bedrock: Using Amazon Bedrock, you can easily experiment with and evaluate top foundation models for your use cases, privately customize them with your data using techniques such as fine-tuning and Retrieval Augmented Generation (RAG), and build agents that execute tasks using your enterprise systems and data sources. 6. Google AI: # Azure Foundry

Azure AI Foundry is Microsoft’s end-to-end platform for building, customizing, deploying, and managing AI applications on Azure. It provides a unified environment for working with foundation models, enterprise data, and production-ready AI tooling. The platform supports Azure OpenAI models such as GPT-4, GPT-4o, and GPT-3.5, enabling advanced generative AI use cases. It also includes Microsoft’s own models like Phi, optimized for efficiency and controllability. In addition, Azure AI Foundry supports leading open models including Llama and Mistral. Teams can compare, evaluate, and select models through a centralized model catalog. The platform enables retrieval-augmented generation (RAG) by securely grounding models in organizational data. Built-in evaluation, monitoring, and safety features help ensure responsible and reliable AI behavior. Azure AI Foundry integrates seamlessly with Azure services for identity, security, and compliance. It is designed to help organizations move from experimentation to scalable, governed AI deployment. To set up an Azure Foundry deployment follow the steps from[ Microsoft Azure OpenAI services](https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/create-resource?pivots=web-portal). It is important to remember that Azure Foundry often has Beta mode and some features must be requested from Microsoft Azure support to unlock the full potential. {% hint style="info" %} Additionally pay attention to the region which you deploy as some features may not be available in certain regions. East US2 currently has the most availability and features along with ChatGPT 5, which allows for using azure search to query your data {% endhint %} Once you have created an Azure Foundry services account and deployed your first model, we will need to grab three values to enter into our system. Navigate to your admin portal/ Navigate to the 'edit company' tab and click Add Azure Deployment under the Azure OpenAI settings.

Here we will need to enter three values: * [x] Your Azure OpenAI endpoint \[https\://\[SOMETHING].openai.azure.com/] * [x] Your Azure OpenAI key * [x] Your Azure OpenAI Deployment name Your OpenAI Endpoint and Key can be found on the [azure portal](https://portal.azure.com/) under the Azure OpenAI service you created and Keys and Endpoints.

Copy and paste the API Key and Endpoint into the respective fields in the AIC Risk Admin Portal.

Next click the 'Add New' button and enter the name of the deployment you created.

This can be found in the [Azure AI Studio](https://oai.azure.com/portal/) under deployments as highlighted below.

### Azure Search Azure search allows you to add a search query to your agent that can access your internal databases and documents. Azure AI search can not be used with assistant option and is only available when the option of assistant is unchecked on an agent. To create an Azure AI search for your data follow the instructions on the Azure AI search documentation on getting started. {% embed url="" %} Once you have created a deployment for your model and added a search to your data through azure click 'Add Azure Search'.

You can add multiple searches but for each one we will need three values: * Search Endpoint https\://{SOMETHING}.search.windows.net/ * Index Name (Found under Indexes on the search service you created) * Endpoint Key (Found under Keys on the search service you created)

# OpenAI To utilize AIC Risk you need to have an OpenAI account and a API Key.

To set up an OpenAI deployment first, create an [OpenAI account](https://platform.openai.com/signup) or [sign in](https://platform.openai.com/login). Next, navigate to the [API key page](https://platform.openai.com/account/api-keys) and "Create new secret key", optionally naming the key. {% hint style="danger" %} **You will only see the Key Once so make sure to copy the key out after you create it.!** {% endhint %} {% hint style="info" %} We recommend you create a new key if you already have an account. In the future OpenAI will allow you to monitor usage by API keys {% endhint %} {% embed url="" %} Navigate to the company tab and click 'Add OpenAI Endpoint' under the OpenAI settings.

Copy and paste the OpenAI key you created for AIC Risk in the[ OpenAI portal](https://platform.openai.com/api-keys). Now you will be able to create agents with OpenAI! # Mistral In order to get a Mistral account with an API key you must first request to join Mistral AI. It usually takes a day for them to approve the account.

The form to join Mistral AI can be found [here](https://mistral.ai/product/). Once you have created an account and have been approved, navigate to your portal and select API Keys or [click here](https://console.mistral.ai/user/api-keys/). [MISTRAL API KEYS](https://console.mistral.ai/user/api-keys/) {% hint style="danger" %} **You will only see the Key Once so make sure to copy the key out after you create it.!** {% endhint %} {% hint style="info" %} We recommend you create a new key if you already have an account. {% endhint %} Navigate to the company tab in the AIC Risk Admin Portal and click 'Add Mistral Endpoint' under the Mistral settings.

{% embed url="" %} Copy and paste the Mistral key you created for AIC Risk in the [Mistral User Portal](https://console.mistral.ai/user/api-keys/). Now you will be able to create agents with Mistral! # Google AI

# LLaMA AI

# AWS Bedrock

# Chat The chat feature allows users to interact with agents in real-time to seek assistance in various topics. This section walks you through initiating a chat and viewing past chats.

# Initiating a Chat Step-by-Step guide to initiating a chat 1. **Accessing the Chat Interface** * Navigate to the application by logging in with your username and password. * Upon successful login, you will be directed to the home screen of the application.

2. **Selecting an agent** * Locate the 'Select Agent' dropdown menu on the left side of the screen. This menu contains the agents currently available to you. * Click on the dropdown menu to expand the list of available chat agents. * Select the agent that best corresponds to the type of chat you wish to initiate. For example, options may include "General Inquiry," "Human Resources," "Sales," etc. * Click on the selected agent to proceed.

3. **Attaching a File (If Applicable)** * Depending on the selected agent, you may have the ability to attach files, such as .pdf or .docx files. * If file attachment is supported for the chosen agent, look for a designated "Attach File" button or icon within the chat interface. * Click on the "Attach File" button/icon to open the file selection dialog. * Browse your device's file system to locate the file you wish to attach. * Select the desired file and click "Open" or "Attach" to include it with your chat message.

4. **Initiating the Chat** * Once you have selected the appropriate chat option and attached any necessary files, proceed to initiate the chat by typing your message in the chat window.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EQjexrMqVDZIrhqO-RDWt_0BwyVPq2ylXeoe4iiXhJgSNA?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=yPTgY4). #### Additional Information: * **Prompt Clearly**: Be sure to communicate your questions and inquiries clearly and concisely within the chat interface. The average response time is approximately 10-15 seconds though some responses may take longer depending on the complexity of the task. # Viewing Chat History Step-by-Step guide to viewing a past chat 1. **Viewing Chat History** * Assuming you are logged in, locate the list of past chats or conversations on the left side of the home screen. * Click on the conversation you wish to view from the list * The chat history will be displayed in the main chat interface, allowing you to review past messages and interactions.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EZkne8Ls5b9DhUMoOXkKsWkBDn4eOW6HbyG3cAwchoHYig?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=fbeX2H). # Dashboards # Creating Dashboards * [x] Only companies AIC Risk Finance can create dashboards customized dashboards. Otherwise you will only have the default dashboards. * [x] To request new dashboard elements or queries[ click here](https://www.aicrisk.com/forum/new-features) to add a new [feature request](https://www.aicrisk.com/forum/new-features) * # Dashboard Elements A dashboard element can be configured to specific Time Frames, Agents or Scanners. Each element has unique functionalities. Each Element starts with 1 of 4 types: * [Top (List of the top items from the query)](/dashboards/dashboard-elements/top-n) * [Charts and Graphs ](/dashboards/dashboard-elements/charts-and-graphs) * [Total (Change over time periods)](/dashboards/dashboard-elements/totals) * [Map (Locations of conversations globally)](/dashboards/dashboard-elements/maps) # Top N Click "Add Item" and select "Top Users". This shows by user number of conversations and flags in the past month.

# Charts and Graphs Click "Add Item" and select "Chart Conversations By Date". This shows a graph of the conversations versus flags within the past month.

Click "Add Item" and select "Chart Total Cost By Date". This shows a graph of the total cost by date within the past month.

# Totals Click "Add Item" and select "Total Conversations Last Month". This chart shows the total conversations within the past month versus the previous period.

Click "Add Item" and select "Total Cost Last Month". This shows a chart of agent prompt costs, completion costs, additional costs, and total costs within the past month.

Click "Add Item" and select "Total Cost By User Last Month". This shows a chart of user prompt costs, completion costs, additional costs, and total costs within the past month by user.

Click "Add Item" and select "Total Flags Last Month". This chart shows the total flags within the past month versus the previous period.

Click "Add Item" and select "Total Issues Closed Last Month". This chart shows the total issues closed within the past month versus the previous period.

Click "Add Item" and select "Total Issues Open Last Month". This chart shows the total issues open within the past month versus the previous period.

# Maps Click "Add Item" and select "Map Of World". This shows a graph of the conversations globally within the past month.

# Monitoring # Adding a User to the Monitor Role This section provides step-by-step instructions for adding a user to the monitoring role within your company. **Adding a User to the Monitoring Role:** To add a user to the monitoring role in our application, follow these steps: 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

2. **Viewing Users:** * Upon accessing the user's section, you will be presented with a list of users registered in the system. * The list may include user details such as email addresses and first and last names.

3. **Filtering Users:** * Use the search bar to narrow down the list of users based on specific criteria such as email address, first name, or last name.

4. **Viewing User Details:** * Click on the user to view detailed information in the user panel. * Locate the 'Roles' section and check the box labeled 'Is Monitor'. * Click save and exit the user panel to save the changes.

**To access open issues:** * Look for the "Open Issues" tab in the menu bar or navigation panel. * Click on the "Open Issues" tab to access issues that need resolving.

**To access closed issues:** * Look for the "Closed Issues" tab in the menu bar or navigation panel. * Click on the "Closed Issues" tab to access issues that have already been resolved.

# Compliance The compliance role within the system grants users access to the compliance section of the platform, where searches can be conducted to ensure adherence to regulatory requirements, internal policies, and ethical standards. By conducting searches, users in this role contribute to the organization's overall compliance efforts and promote a culture of accountability and transparency. **Responsibilities:** **Access Compliance Section:** Utilize authorized access to the compliance section to conduct searches and investigations related to flagged events or compliance issues.

**Search and Analysis:** Conduct searches to identify past conversations related to specific keywords, topics, or flagged events. Analyze conversation transcripts to identify any compliance issues or irregularities.

# Adding a User to the Compliance Role This section provides step-by-step instructions for adding a user to the compliance role within your company. **Adding a User to the Compliance Role:** To add a user to the compliance role in our application, follow these step 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

3. **Filtering Users:** * Use the search option to narrow down the list of users based on specific criteria such as email address, first name, or last name.

4. **Viewing User Details:** * [Click on the user to view detailed information in the user panel.](#user-content-fn-1)[^1] * Locate the 'Roles' section and check the box labeled 'Is Compliance'.

* Click the save button and exit the user panel to save the changes.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EWRr-e5VvTBLkZRVvaQE0MAB1phKsrmSORGm654MAHXcog?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=5aesUB). **Note:** Ensure that you have the necessary permissions to access and manage user roles. Exercise caution when performing actions that affect user accounts to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage user roles. [^1]: # Creating a Search This section provides step-by-step instructions for creating a compliance search within your company. **Creating a Search:** To create a search in our application, follow these steps: 1. **Navigating to the Compliance Section:** * Sign in to your account using your credentials. * Click on the menu icon located in the top-left corner of the screen. * From the menu, select the "Compliance" option if available. This will direct you to the compliance section of the platform.

2. **Accessing the New Search:** * Within the Compliance section, locate and click on the "New Search" tab. This tab should provide you with the steps to proceed with a new search.

3. **Creating the New Search:**

**There are 5 steps to creating a new search:** 1. **Reason:** Give the search a name and describe the reason you are performing the search.

2. **Users:** Select the specific users in your company whose activity you are searching for from the list.

3. **Search Text:** Enter the key words to search for in the text box. You can enter multiple key words separated by commas.

4. **Date Range:** Select a date range to search within.

5. **Finish:** Preview the results and click the button labeled 'GO' to view. A new tab labeled with the search's name will appear containing the results.

**Viewing the Results:** * Use the search option to narrow down the list of results based on specific criteria such as the prompt, user, agent name, or date created.

* Click on the desired result to view detailed information in the results panel.

**Note:** Ensure that you have the necessary permissions to create a search. Searches are strictly used for compliance purposes and all searches are recorded in the system. Exercise caution to avoid unintended consequences. #### Additional Information: * **Compliance Privileges**: Please note that only users with compliance privileges can access the compliance section and manage searches. # Consuming the Results Once you've accessed the compliance section of the platform, you can view any past search in the company and analyze its results. **Analyzing Search Results:** 1. Use the search feature in the 'My Activity' or 'All' tab to narrow down the list of searches based on the search name/title, user who created the search, or search text. 2. Click the desired search to view its results. 3. Click a search result to view its detailed information in the results panel. 4. The results panel contains two sections to review: #### Risk Results: * **Text Prompt:** Displays the prompt that was sent to the agent.

* **Conversation:** Click the button labeled 'Go To Conversation' to view the full chat transcript.

* **Language:** Describes the language in which the chat was written.

* **Country:** Displays the country in which the conversation was initiated

* **IP Address:** Displays the IP address of the user who initiated the conversation.

* **Is In European Union:** Displays whether or not the user who initiated the conversation is a part of the European Union.

* **Flagged:** Displays whether or not the conversation has been reviewed and flagged in the Monitoring section of the application.

#### Risk Reports: This section displays the scanner violation(s) that occurred within the conversation. To learn more about our proprietary scanners and their functionalities, click [here](/admin/ai-agents/adding-a-new-agent/advanced-settings/scanners).

# Admin This section provides step-by-step instructions for accessing the admin panel. In the admin panel you can manage agents, company information, users, groups, scanner defaults, and logs. **Accessing the Admin Panel:** To access the admin panel, follow these steps: 1. On the Home Screen, locate the dropdown menu in the top-left corner of the interface.

2. Click on the dropdown menu to view its options.

3. Within the dropdown menu, locate and select the "Admin" option. This action will redirect you to the Admin Panel.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EUNBzjUmO6xLsAU_fXx43vwB1mWBtlTTbLYHMAp6ws6V-Q?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=S4fA1x). #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel. # AI Agents This section provides step-by-step instructions for accessing the AI agent section of the platform. Learn how to create, edit, and delete agents and templates. **Accessing AI Agents:** To access your company's agents in our application, follow these steps: 1. **Navigating to the Agents Section:** * Log in to the admin panel using your credentials. * Look for the "Agents" tab in the menu bar or navigation panel. It should be the first section you see when opening the admin panel.

2. **Viewing Agents:** * Upon accessing the agents section, you will be presented with a list of agents registered in the system. * The list may include agent details such as name, description, and agent engine or type.

3. **Filtering Agents:** * Use the search bar option to narrow down the list of agents if necessary to help locate the one you are looking for. 4. **Viewing Agent Details:** * Click on the agent to view detailed information in the agent editing panel.

To watch a video demonstration, click[ ](https://airiskinc.sharepoint.com/:v:/s/ArtificialIntelligenceRisk/EbfiL8tXotZIkfigM0h4u7sBXCNNW-hVMSC9L9565EP8CQ?e=sPz541)[here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EbPzssG9DbVEj94xAUtR31YBYBtN80Ly8IUeCuId0ArxQA?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=CauFNK). **Note:** Ensure that you have the necessary permissions to access and manage agents. Exercise caution when performing actions that affect user accounts to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage agents. # Adding a New Agent This section provides step-by-step instructions for adding a new agent. **Adding a New Agent:** To add a new agent to your company, follow these steps: 1. **Navigating to the Agents Section:** * Log in to the admin panel using your credentials. * Look for the "Agents" tab in the menu bar or navigation panel. It should be the second option. * Click on the "Agents" tab if necessary to access the agent's section.

2. **Accessing the Create New Agent Panel:** * Within the Agents section, locate the button labeled "Add New Agent".

* Click on the "Add New Agent" button to access the Create New Agent panel.

3. **Creating the New Agent:** * Within the Create New Agent panel, enter the new agent details. There are two sections to review. 1. General Settings - To learn more about general settings click [here](/admin/ai-agents/adding-a-new-agent/general-settings).

2. Advanced Settings - To learn more about advanced settings click [here](/admin/ai-agents/adding-a-new-agent/advanced-settings).

4. **Finalizing the New Agent:** * After entering the required information, review it for accuracy. * Once satisfied, click save and exit the create new agent panel to save the changes.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/Eb80tfo32adKtGsFXw7H3KgBmeMTWxU6xCRv1CJYU_6SOA?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=O57zwk). #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage agents. # General Settings The general settings are for configuring and customizing the agent's identity, behavior, and operational parameters. By providing clarity and guidance within this section, users can effectively tailor the agent to meet their specific needs and objectives.

## **General settings include:** ### **Agent Name:** * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the agent. * *Usage:* Enter a concise and descriptive name that succinctly identifies the agent within the system. This name should facilitate easy recognition and differentiation among multiple agents.

### **Description:** * *Input Field Type:* Text Area * *Description:* Users can provide a brief overview or description of the agent, detailing its purpose, capabilities, or any relevant information. * *Usage:* Utilize this space to offer insights into the agent's role, functionality, or any specific instructions pertinent to its usage. A clear and informative description enhances understanding and facilitates effective utilization of the agent.

### **Default Prompt:** * *Input Field Type:* Text Area * *Description:* The Default Prompt serves as the initial message or directive presented to the agent, providing guidance on its intended use and instructions on how to handle user prompts. * *Usage:* Craft a concise and clear directive that outlines the agent's purpose, directed use, and expected behavior when interacting with users. The Default Prompt sets the tone for conversations and instructs the agent on how to respond to user inquiries or prompts effectively. It serves as a foundational element in guiding the agent's interactions and ensuring alignment with its designated role and objectives.

### **External Agent Engine:** * *Input Field Type:* Dropdown Menu * *Description:* Users can select from a list of available external agent engines, which determine the underlying technology or platform powering the agent's functionality. * *Usage:* Choose the appropriate external agent engine based on specific requirements, preferences, or compatibility considerations. The selected engine influences the agent's capabilities, performance, and integration possibilities.

# Agent Engine Properties Learn about agent engine properties including OpenAI, Mistral, and Gemini.

# Azure OpenAI https\://azure.microsoft.com/en-us/products/ai-services/openai-service/

{% embed url="" %} # OpenAI https\://platform.openai.com/docs/overview

## These are OpenAI models available on our platform. Each one offers unique settings and abilities: **`gpt-3.5-turbo:`** Currently points to `gpt-3.5-turbo-0613`. The `gpt-3.5-turbo` model alias will be automatically upgraded from `gpt-3.5-turbo-0613` to `gpt-3.5-turbo-0125` on February 16th, 2024. Data up to Sep 2021. * [x] Is Assistant * [x] Code Interpreter * [x] User Can Upload Files **`gpt-3.5-turbo-1106:`** GPT-3.5 Turbo model with improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more. Returns a maximum of 4,096 output tokens. Data up to Sep 2021. * [x] Is Assistant * [x] Code Interpreter * [x] File Retrieval * [x] User Can Upload Files **`gpt-4:`** Currently points to `gpt-4-0613`. Data up to Sep 2021. * [x] Is Assistant * [x] Code Interpreter * [x] User Can Upload Files **`gpt-4-1106-preview:`** GPT-4 Turbo model featuring improved instruction following, JSON mode, reproducible outputs, parallel function calling, and more. Returns a maximum of 4,096 output tokens. This is a preview model. Data up to Apr 2023. * [x] Is Assistant * [x] Code Interpreter * [x] File Retrieval * [x] User Can Upload Files **`gpt-4-vision-preview:`** GPT-4 with the ability to understand images, in addition to all other GPT-4 Turbo capabilities. Currently points to `gpt-4-1106-vision-preview`. Data up to Apr 2023. * [x] Is Assistant * [x] Code Interpreter * [x] User Can Upload Files ## **Code Interpreter** Code interpreter is a function of OpenAI which allows assistants to write python code to generate charts or other mathematical calculations for more information follow the link below. {% embed url="" %} ## **Knowledge Retrieval** Knowledge Retrieval allows the assistant to work with data from outside the model including documents that are uploaded to the agent. This must be enabled to allow file uploading. For more information follow the link below. {% embed url="" %} # Mistral https\://docs.mistral.ai/

## These are Mistral models available on our platform. Each one offers unique abilities: **`open-mistral-7b(mistral-tiny-2312):`**Mistral 7B is the first dense model released by Mistral AI. At the time of the release, it matched the capabilities of models up to 30B parameters. To learn more about its capabilities, click [here](https://mistral.ai/news/announcing-mistral-7b/). **`mistral-small-latest(mistral-small-2402):`**Mistral Small is the ideal choice for simple tasks that one can do in bulk - like Classification, Customer Support, or Text Generation. **`mistral-medium-latest(mistral-medium-2312):`**Mistral Medium is ideal for intermediate tasks that require moderate reasoning - like data extraction, summarizing a document, writing a job description, or writing product descriptions. Mistral Medium strikes a balance between performance and capability, making it suitable for a wide range of tasks that only require language transformation. For example, Mistral Medium can write an email. {% embed url="" %} # Google AI

# LLaMA AI

# AWS Bedrock

# Creating an Agent with AWS Bedrock Amazon Web Services (AWS) has introduced new tools that make it easier for developers to build AI agents capable of reasoning, decision-making, and interacting with other systems. **Video walkthrough** [**here**](https://airiskinc.sharepoint.com/:v:/s/ArtificialIntelligenceRisk/EbN_lrwO8lJFgX-yJDUsD7gBYimugqGDTX3NQOt3fWw1Ew?e=avyzEu\&nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D)**.** **1. Go to the Admin Panel**\ Navigate to the admin section of your dashboard to access agent management tools. **2. Click “Add New Agent”**\ Locate and select the **“Add New Agent”** button to begin creating a new agent profile.

**3. Enter the Agent Name, Description, and Default Prompt. See** [**General Settings**](/admin/ai-agents/adding-a-new-agent/general-settings)**.**

**4. (Optional) Add a Suggested Question**\ In the **Suggested Questions** field, click on "Add Suggestions". Then, add a greeting and description, click the checkmark, and save. You may add as many suggestions as you'd like.

**5. Select the Agent Engine: AWS Bedrock**\ Before saving, make sure to select **AWS Bedrock** as the agent engine.

**6. Choose the Model Under AWS Bedrock Settings**\ Under **AWS Bedrock Settings**, choose the appropriate model you want the agent to use. **7. Check the “Enable File Search” Box**\ To allow your agent to retrieve information from uploaded documents, make sure to check the **“Enable File Search”** box. **8. Choose an Existing Bucket or Create a New One**\ Type in a bucket name to use either an existing one or create a new one. ***Bucket*****:** is a place to store files. It’s safe for different agents to share the same bucket name, as they won’t be able to access or read each other’s files. Users can use an existing bucket or create a new bucket. The buckets are accessible in the AWS S3 Console.

**9. (Optional) Check the “User Can Upload Files” Box**\ If you want end users to be able to upload their own documents via the chat screen for the agent to use, check the **“User Can Upload Files”** box.

**10. Click “Add Files” to Upload Documents for the Agent**\ Use the **“Add Files”** button to upload any documents you want the agent to access. All files assigned to that agent will be accessible across all conversations for a given user.

**11. Add Any Data Functions You Want the Agent to Use**\ Scroll to the **Add** **Data** **Functions** section and select any pre-configured functions you’d like this agent to access. These can include things like Web Functions, Custom API Functions, and Custom Database Functions.

**12. See** [**Advanced Settings**](/admin/ai-agents/adding-a-new-agent/advanced-settings)**.** # Advanced Settings The advanced settings section of the agent configuration interface provides users with additional control and customization options to fine-tune the behavior, security, and integrations of the agent. ## **Advanced settings include:**

### **Usage:** * *Description:* This setting allows the user to specify the **Max Messages** and **LLM Temperature** for the agent. **Max Messages** refers to the number of messages the agent will refer back to in the chat to create a response. * *Usage:* Enter the values that are appropriate for the agent you are creating. To learn more about **LLM Temperature** click [here](https://deepchecks.com/glossary/llm-parameters/).

### **Scanners:** * *Description:* This field allows the user to specify which unique scanners the agent will use. * *Usage:* Enable the AI Scanners appropriate for the agent you are creating. The seven proprietary scanners are: [Toxicity](/admin/agent-scanner-defaults/toxicity), [Personal Information](/admin/agent-scanner-defaults/personal-information), [Topic](/admin/agent-scanner-defaults/topic-scanner),[ Prompt Injection Detection](/admin/agent-scanner-defaults/prompt-injection-detection), [Regex](/admin/agent-scanner-defaults/regex-scanner), [Geographic Gating](/admin/agent-scanner-defaults/geographic-gating), and [Language Detection](/admin/agent-scanner-defaults/language-detection).

### **API Keys:** * *Description:* This field allows the user to view the company's AIR Key and external agent engine key. * *Usage:* Enter an updated external agent engine key if the present one is outdated.

### **Files:** * *Description:* This field allows the user to specify which file types are allowed to be uploaded to the agent. This setting will only become visible if File Uploads are enabled. * *Usage:* To enable File Uploads, make sure you select an agent engine and model that supports this feature. You must also enable the ***Is Assistant*** and ***Code Interpreter*** options to tell the agent it can analyze the attached data. The screenshot below shows the OpenAI agent engine selected utilizing its`gpt-4`model:

# Usage The usage settings empower users to customize the agent's behavior and response generation process according to specific requirements and preferences. By adjusting settings such as Max Messages and LLM Temperature, users can optimize the agent's performance, enhance conversational quality, and tailor interactions to suit diverse use cases and scenarios.

# Max Messages This parameter caps the amount of tokens sent to the third party agent and can reduce costs. The total messages the agent will have access to with each prompt. Note the initial default (system) prompt will always be loaded into the chat to keep context.

**Max Messages:** * *Input Field Type:* Numeric Input * *Description:* The Max Messages setting defines the maximum number of messages the agent can refer back to in the chat history during conversations. * *Usage:* Specify an appropriate limit for the maximum number of messages the agent can access or refer to within the chat history. The default value is 20 messages. This setting influences the depth and context available to the agent during interactions with users, impacting the quality and relevance of responses. # LLM Temperature Temperature is a physical property that measures the degree of hotness or coldness of an object or substance. In the context of Large Language Models (LLMs), temperature is a hyperparameter used during the generation of text. A lower temperature value makes the output more deterministic and less likely to explore different possibilities, while a higher temperature value makes the output more exploratory and more likely to generate diverse responses.

**LLM Temperature:** * *Input Field Type:* Numeric Input * *Description:* LLM Temperature determines the degree of randomness or creativity in the agent's responses, controlling the diversity and variability of generated text. * *Usage:* Adjust the LLM Temperature to regulate the level of novelty and spontaneity in the agent's output. Higher temperatures encourage more exploratory and diverse responses, while lower temperatures prioritize coherence and relevance. To learn more about LLM temperature click [here](https://deepchecks.com/glossary/llm-parameters/). # Scanners The scanners section within the agent configuration interface equips users with powerful tools for monitoring, filtering, and analyzing incoming data to ensure compliance, security, and relevance in agent interactions. Scanners are essential components that enhance the agent's capability to detect and respond appropriately to various types of content and context.

# AI Scanners ### **AI Scanners:**

1. [**Toxicity Scanner**](/admin/agent-scanner-defaults/toxicity) * *Description:* The Toxicity Scanner identifies and filters out potentially harmful or offensive content from user input or generated responses. * *Usage:* By analyzing the text for indicators of toxicity or inappropriate language, the Toxicity Scanner helps maintain a safe and respectful conversational environment.

2. [**PII (Personally Identifiable Information) Scanner**](/admin/agent-scanner-defaults/personal-information) * *Description:* The PII Scanner detects and masks personally identifiable information to protect user privacy and comply with data protection regulations. * *Usage:* By recognizing sensitive information such as names, addresses, or contact details, the PII Scanner ensures that such data is handled appropriately and securely.

3. [**Topic Scanner**](/admin/agent-scanner-defaults/topic-scanner) * *Description:* The Topic Scanner categorizes user input and facilitates topic-based response generation for improved relevance and coherence. * *Usage:* By analyzing the semantic context and identifying key topics within user messages, the Topic Scanner enables the agent to generate responses that align with the ongoing conversation.

4. [**Prompt Injection Detection**](/admin/agent-scanner-defaults/prompt-injection-detection) * *Description:* Prompt Injection Detection identifies and mitigates attempts to manipulate or inject unauthorized prompts into agent interactions. * *Usage:* By monitoring for anomalies or unexpected changes in conversation dynamics, Prompt Injection Detection helps maintain the integrity and authenticity of interactions.

# General Scanners ### **General Scanners:**

1. [**Regex Scanner**](/admin/agent-scanner-defaults/regex-scanner) * *Description:* The Regex Scanner utilizes regular expressions to identify and extract patterns or specific information from user input or generated responses. * *Usage:* By defining custom patterns or rules, the Regex Scanner enables users to extract relevant data or trigger specific actions based on predefined criteria.

2. [**Geographic Gating**](/admin/agent-scanner-defaults/geographic-gating) * *Description:* Geographic Gating restricts access or modifies agent behavior based on the user's geographic location or specified regions. * *Usage:* By applying geographical filters, users can tailor the agent's responses or restrict access to certain features or content based on user location.

3. [**Language Detection**](/admin/agent-scanner-defaults/language-detection) * *Description:* Language Detection identifies the language used in user input or generated responses, enabling multilingual support and language-specific processing. * *Usage:* By detecting the language of incoming messages, the Language Detection Scanner enables the agent to provide appropriate responses and language-specific services.

# API Keys Within the context of the agent configuration interface, API keys play a crucial role in facilitating secure communication and interaction between the agent and external systems or services.

### **AIR Key:** The AIR Key serves as an authentication token specifically designed for accessing the AIR platform. This key is provided at sign-up and enables communication between agents and external agent engines. It allows licensed users to access the platform and leverage advanced natural language processing and generation algorithms for conversational interactions.

### **Agent Engine API Key:** The Agent Engine API Key is a unique authentication token associated with the agent engine instance deployed or utilized by the user. This key authenticates access to the specific agent engine, ensuring secure communication and integration with external applications, platforms, or services. It serves as a protective measure to restrict unauthorized access and safeguard sensitive data exchanged between the agent and external systems. #### **Security Considerations:** API keys are sensitive credentials that grant access to privileged resources and functionalities. As such, it's essential to handle them with care and adhere to best practices for security: 1. **Secure Storage:** API keys should be stored securely, preferably in encrypted format, to prevent unauthorized access or theft. 2. **Restricted Access:** Limit access to API keys only to authorized personnel or systems with a legitimate need for interaction. 3. **Rotation and Renewal:** Regularly rotate API keys and revoke access for outdated or compromised keys to mitigate security risks. 4. **Audit Trails:** Maintain audit trails to monitor API key usage and detect any suspicious or unauthorized activities. By adhering to these security measures and best practices, users can ensure the integrity, confidentiality, and availability of their API keys, thereby enhancing the overall security posture of the agent and associated systems. # Files Specify which file types are allowed to be uploaded to the agent as well as whether or not the enabled scanners apply to those files. ### Files Settings:

To restrict a file type from an agent, click the edit icon to open the 'Allowed File Types' panel. Uncheck the box(s) of the file type(s) you want to exclude:

# Editing an Agent This section provides step-by-step instructions for editing an existing agent. **Editing an Agent:** To edit an existing agent in your company, follow these steps: 1. **Navigating to the Agents Section:** * Log in to the admin panel using your credentials. * Look for the "Agents" tab in the menu bar or navigation panel. It should be the first option. * Click on the "Agents" tab if necessary to access the agent's section.

2. **Selecting the Agent to Edit:** * Within the "Agents" section, locate the agent you want to edit from the list of available agents. * Use the search bar to narrow down the list of agents based on specific criteria such as agent name, description, and type. * Click on the agent's name to access it's editing panel.

3. **Editing the Agent:** * In the agent's editing panel, make any appropriate changes needed. * After making the desired changes in the editing panel, review it for accuracy. * Once satisfied, click save and exit the editing panel to save the changes.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EcI0vx_c4F1Km82NPWLiS8QBmkwtHEYTbG8npyUPCrs2xg?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=Tbibxr). #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage agents. # Deleting an Agent This section provides step-by-step instructions for deleting an agent from your company. **Deleting an Agent:** To remove an agent from your company, follow these steps: 1. **Navigating to the Agents Section:** * Log in to the admin panel using your credentials. * Look for the "Agents" tab in the menu bar or navigation panel. It should be the first option. * Click on the "Agents" tab if necessary to access the agent's section.

2. **Selecting the Agent to Delete:** * Within the "Agents" section, locate the agent you want to delete from the list of accessible agents. * Click on the agent's name to access it's editing panel.

3. **Initiating the Deletion Process:** * Within the editing panel, look for a button at the bottom labeled "Delete Agent" or similar. * Click on the "Delete Agent" button to initiate the deletion process.

4. **Confirming Agent Deletion:** * A confirmation prompt may appear to ensure you intend to delete the agent. * Review the confirmation message carefully.

5. **Finalizing Agent Deletion:** * Confirm the deletion action if prompted. * Once confirmed, the agent will be permanently deleted from the system. To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EXjfJwJtLRhBjQCkx2bcpWkBsLQlKSj0aQvmZXeRCmZV6A?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=EIqvpt). **Note:** Deleting an agent is irreversible and will result in removing user's access to the deleted agent. Ensure that you intend to delete the agent and review the action carefully before confirming. **Note:** All interactions within the platform are recorded for compliance purposes. These records will remain in the system even after an agent is removed. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage agents. # Adding from a Template This section provides step-by-step instructions for adding an agent from an existing template. **Adding from a Template:** To add an agent from an existing template, follow these steps: 1. **Navigating to the Agents Section:** * Log in to the admin panel using your credentials. * Look for the "Agents" tab in the menu bar or navigation panel. It should be the first option. * Click on the "Agents" tab if necessary to access the agent's section.

2. **Accessing the Template Selector Window:** * Within the "Agents" section, locate the button labeled "Add From Template". * Click on the "Add From Template" button to access the template selector window.

3. **Creating the New Agent:** * In the template selector window, you are presented with the templates currently created for your company.

* Use the filtering options to narrow down the list of templates based on specific criteria such as agent engine type, category, and department.

* Click the button labeled 'Create' associated with the template you wish to use to access to the editing panel. 4. **Editing the Agent:** * In the template editing panel you can make any appropriate changes needed for the new agent. * After making the desired changes in the editing panel, review it for accuracy. * Once satisfied, click save and exit the editing panel to save the changes. * Refresh the page to see the new agent created in the list of AI Agents available to your company. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage agents. # Company This section provides step-by-step instructions for accessing the company management section of the platform. This includes accessing the billing portal and managing external agent provider settings. **Accessing the Company Section:** To access the company management section, follow these steps: 1. **Navigating to the Company Section:** * Log in to the admin panel using your credentials. * Look for the "Edit Company" tab in the menu bar or navigation panel. * Click on the "Edit Company" tab to access the company section.

2. **Viewing your Company:** * Upon accessing the "Edit Company" section, you will be presented with a button labeled 'Billing Portal' and two sections labeled 'General' and 'External Agent Provider Endpoints.' * The 'General' section shows details such as your company name and user license count

* The 'External Agent Provider Endpoints' section shows details such as provider keys and access to each provider's billing portal.

* The 'Billing Portal' button allows access to the Stripe Payment Portal to manage the company's subscription and add license keys. 3. **Viewing External Provider Details:** * Click on the dropdowns to view detailed information about each external agent provider endpoint. * These details include agent provider keys, Microsoft Azure OpenAI deployment information, and access to each provider's billing portal. **Note:** Ensure that you have the necessary permissions to access and manage company settings. Exercise caution when performing actions that affect company access to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage company settings. # General Information The "General Information" section provides essential details about the company associated with the account. **General Information:** 1. **Company Name**: This field displays the name of the company. Users with appropriate permissions can edit this field as needed. The company name is important for identification purposes and appears in various contexts within the platform.

2. **Total User Licenses**: This section indicates the number of user licenses allocated to the company. User licenses determine the number of individuals who can access and utilize the platform within the company's account. Depending on the subscription or licensing model, the number of user licenses may vary. Users with administrative privileges typically manage and allocate these licenses within the system.

3. **Speech to Text**: This checkbox allows you to configure if you would like to allow users to be able to use text to speech. By default turned off. (Note: this function is using speech to text webapi through your default browser to execute and may not comply with company security policy)

**Usage** * **Editing Company Name**: To edit the company name, click on the editable field and input the desired name. Ensure the new name accurately reflects the company's identity.

* **Managing User Licenses**: User licenses are finite resources allocated to the company. Administrators can manage these licenses by adding or removing users as needed. It's crucial to monitor license usage to ensure compliance with the company's subscription plan and avoid overages or underutilization. If you need to increase you license amount contact **Default EntraID Group** **Default**: Any user who has access to this entra id group can log into the system and will be assigned all agents in the Default group created in the application

**Best Practices** 1. **Accurate Company Name**: Keep the company name updated to reflect any changes in the organization's branding or legal entity. 2. **Optimizing License Allocation**: Regularly review user licenses to reallocate them based on evolving organizational needs. Remove licenses from inactive users and assign them to new team members to maximize resource utilization. 3. **Monitoring Usage**: Maintain awareness of license usage trends to anticipate future needs accurately. Consider upgrading or adjusting subscription plans to accommodate growing user counts or changing usage patterns. **Security Considerations** Ensure that only authorized personnel have access to manage company information and user licenses. Protect sensitive data related to the company's identity and subscription details from unauthorized access or modification. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage company settings. # External Agent Provider Settings The "External Agent Provider Settings" section provides a centralized location for configuring and managing settings related to external agents offered by the platform. **External Agent Provider Settings:** By offering specific settings tailored to each external agent, users can seamlessly integrate, configure, and manage interactions with external services such as OpenAI, Azure OpenAI, and Mistral. Below are the details of the settings available for each external agent: 1. **OpenAI Settings**: * **Keys**: Users can view and manage the API key associated with the OpenAI external agent. This includes the option to edit or update for seamless integration with the platform. * **Billing Portal Access**: Users have the ability to access the billing portal directly from the platform, enabling them to manage billing-related tasks such as viewing invoices, updating payment methods, and managing subscription plans.

2. **Azure OpenAI Settings**: * **Azure OpenAI Endpoint String**: Users can input and manage the Azure OpenAI Endpoint string, which serves as the endpoint for interacting with Azure OpenAI services. * **Deployment Settings**: Users have the option to configure deployment settings for Azure OpenAI. This includes the ability to add new deployments, enabling users to deploy models and manage their lifecycle effectively. * **Azure Search**: Users can add Azure Search functionality, allowing for advanced search capabilities within the Azure OpenAI deployment environment.

3. **Mistral Settings**: * **Keys**: Similar to OpenAI, users can view and manage the API key associated with the Mistral external agent. This includes the option to edit or update for seamless integration with the platform. * **Billing Portal Access**: Users have access to the billing portal directly from the platform, enabling them to manage billing-related tasks such as viewing invoices, updating payment methods, and managing subscription plans.

**Usage** * **Accessing Settings**: To access the settings for a specific external agent, click on the corresponding dropdown menu within the "External Agent Provider Settings" section. This will reveal the settings options available for that particular external agent. * **Editing/Updating Keys**: Users can edit or update these keys directly within the settings menu for each external agent. Ensure that keys are accurately configured to enable seamless integration and interaction with external agent services. * **Configuring Deployment Settings**: When working with Azure OpenAI, users can configure deployment settings such as endpoint strings and deployment management options to streamline the deployment and management of models within the Azure environment. **Security Considerations** * Safeguard API keys and other sensitive information associated with external agents to prevent unauthorized access and misuse. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage company settings. # Custom APIs This section provides step-by-step instructions for accessing the Custom APIs section of the platform. Learn how to create, edit, and delete custom APIs. **Accessing Custom APIs:** To access your custom APIs in our application, follow these steps: 1. **Navigating to the Custom API Section:** * Log in to the admin panel using your credentials. * Look for the "Custom APIs" tab in the menu bar or navigation panel. It should be the fifth section you see when opening the admin panel.

2. **Viewing Custom APIs:** * Upon accessing the custom API section, you will be presented with a list of APIs registered in the system. * The list will include API name and function description.

3. **Filtering APIs:** * Use the search bar to narrow down the list of APIs, if necessary, to help locate the one you are looking for.

4. **Viewing Custom API Details:** * Click on the API name to view detailed information in the Custom API editing panel.

**Note:** Ensure that you have the necessary permissions to access and manage APIs. Exercise caution when performing actions that affect user accounts to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage custom APIs. # OAuth2 API OAuth2 allows you to use user credentials to hit apis utilizing the users login to retrieve data ### Setup In order to setup OAuth2 connections to third party APIs you will need to have the following values entered into your key value. 1. ### Oauth2--\[API-NAME]--ClientID 2. ### Oauth2--\[API-NAME]--ClientSecret 3. ### Oauth2--\[API-NAME]--TokenURL 4. ### Oauth2--\[API-NAME]--Scopes 5. ### Oauth2--\[API-NAME]--RefreshTokenURL 6. ### Oauth2--\[API-NAME]--AuthURL ## EXAMPLE ZOOM Transcripts: To create an oauth connection to zoom you will first need to create a new Zoom application with access to the correct scopes by going to and create a new application. For this example the API Name will be Zoom and therefore your redirect url will use the following callback https\://\[Your AI RIsk Domain]/callback/Zoom {% hint style="info" %} Note that the callback will allways use the format of callback/API Name The api name is also the prefix in the configuration of your secrets {% endhint %} ### Add the following secrets to your key store on azure ### Oauth2--Zoom--AuthURL = ### Oauth2--Zoom--ClientID = \[Your Client ID] ### Oauth2--Zoom--ClientSecret = \[Your Client Secret] ### Oauth2--Zoom--Scopes = cloud\_recording:read:list\_user\_recordings cloud\_recording:read:list\_user\_recordings:admin ### Oauth2--Zoom--TokenURL = # Adding a New API The following 4 sections (this one included) will provide you step-by-step process to create your very own custom API! **Adding a New API:** To add a new API to your company, follow these steps: 1. **Navigating to the Custom API Section:** * Log in to the admin panel using your credentials. * Look for the "Custom APIs" tab in the menu bar or navigation panel. It should be the fifth option. * Click on the "Custom APIs" to access the API section.

2. **Accessing the Create New API Panel:** * Within the Custom API's section, locate the button labeled "Add New API". * Click on the "Add New API" button to access the Create New API panel.

# API Information How to Fill Out API Information The general settings are for configuring and customizing the agent's identity, behavior, and operational parameters. By providing clarity and guidance within this section, users can effectively tailor the agent to meet their specific needs and objectives.

### **API Name:**

* *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs. ### **Call Type:** * *Input Field Type:* Dropdown Menu (we offer three differed request methods for APIs)

{% tabs %} {% tab title="Get" %} Used to RETRIEVE data from a resource {% endtab %} {% tab title="Post" %} Used to CREATE a new resource {% endtab %} {% tab title="Put" %} Used to UPDATE an existing resource {% endtab %} {% endtabs %} * *Description:* pick the specific request method used for the Custom API that they are creating as each type has a different function * *Usage:* Choose the correct request method in order for you API to be able to retrieve the correct data or else the test API feature at the end will not work. The website whose API you are trying to implement will most likely describe which of the three it is. ### **Response Type:**

* *Input Field Type:* Dropdown Menu {% tabs %} {% tab title="Json" %} JSON (JavaScript Object Notation) used for data exchange between systems, APIs, and databases. Represents data as objects (key-value pairs), arrays, and primitive values (strings, numbers, Booleans, null) {% endtab %} {% tab title="Stream" %} Java streams allow operations such as filtering, mapping, reducing, and sorting data from collections, arrays. NOT RELATED TO JSON, instead streaming data within Java applications. Specialization: there are specialized streams for primitive types *IntStream, LongStream, DoubleStream* {% endtab %} {% endtabs %} * *Description:* referring to the format or structure in which an API communicates data back to the requester for data. JSON and Stream are the two retrieval data types offered on our platform. * *Usage:* Choose the correct request method so that when later testing the Custom API it returns the correct information and in the correct format. JSON is readable and represents data such as objects, arrays, and primitive values while Java Streams are efficient and powerful for data manipulation. ### **Use OAuth2:**

* *Input Field Type:* Check Box * *Description:* authorization framework that enables users to safely share their data better between different applications. * *Usage:* Eliminates the need for shared user credentials and provides simple, well-defined authorization flows. * *Box*: Used to identify a user in the Box API, a common server-side authentication method ### **API Endpoint:**

* *Input Field Type:* Text Field * *Description:* User must enter the API endpoint link without parameters as that will be entered later in either section of "API Header", "API Endpoint Parameters", or "API Endpoint Parameters". * *Usage:* Connect the appropriate link to the text field in order to successfully connect to the correct wed database or server, for desired data to be later used by agent using the API function. ### **Function Prompt:**

* *Input Field Type:* Text Area * *Description:* User will write out in the detail what the use of the API will be such as what data it retrieves and what questions will prompt the assigned agent to use this API call. * *Usage:* Utilized by the Agent assigned to the Custom API to decide when to utilize the API function to retrieve specific information and datasets to answer a question posed by the user. # API Headers How to Fill Out API Headers

### Header Name: * *Input Field Type:* Text Field * *Description:* * *Usage:*

### Value: * *Input Field Type:* Text Field * *Description:* * *Usage:*

# API Endpoint Parameters How to fill out Endpoint Parameters ### Parameter Name: * *Input Field Type:* Text Field * *Description:* * *Usage:*

### Value: * *Input Field Type:* Text Field * *Description:* * *Usage:*

### Description: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

### Required: * *Input Field Type:* Check Box * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

### # API Query Parameters How to Fill Out API Query Parameters ### Parameter Name:

* *Input Field Type:* Text Field * *Description:* This field allows the user to name one of the parameters for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

### Value: * *Input Field Type:* Text Field * *Description:* This field allows the user to declare the value of the parameter, if it requires a user's input leave it blank, if the value doesn't change input the unchanged value. * *Usage:* The input in this value field will query the correct information from the dataset of the API and return desired information.

### Description: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify the exact format of needed input for parameter (Ex. Ticker for Company or YYYY-MM-DD). * *Usage:* Enter a concise and descriptive description so that the agent assigned to the Custom API is able to ask the user for the correct further information if needed.

### Required: * *Input Field Type:* Check Box * *Description:* This field allows the user to declare the parameter for the API required or not. Usually, will always be required unless it is the token which where a value will already have an input. * *Usage:* Double check with the API implementation of the Custom API so that the agent utilizing the Custom doesn't crash when trying to query for information it cannot find.

# Body Parameters When doing a POST API request method, it requires user to fill out the Body Parameters section to be filled out, this is how to do so ### GraphQL GraphQL is a type of query language and server-side runtime for APIs. It prioritizes giving clients exactly the data they request and no more, making APIs fast, flexible, and developer friendly. Is not tied to any specific database or storage engine and is backed by existing code and data.

#### Body: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

#### Parameter Name: * *Input Field Type:* Text Field * *Description:* This field allows the user to define a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

#### Value: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

#### Description: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

#### Required: * *Input Field Type:* Check Box * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

*** ### JSON

#### Parameter Name: * *Input Field Type:* Text Field * *Description:* This field allows the user to specify a unique name for the Custom API. * *Usage:* Enter a concise and descriptive name that succinctly identifies the API within the system. This name should facilitate easy recognition and differentiation among multiple APIs.

# Editing a Custom API This section provides step-by-step instructions for editing an existing Custom API. **Editing a Custom API:** To edit an existing Custom API in your company, follow these steps: 1. **Navigating to the Custom API Section:** * Log in to the admin panel using your credentials. * Look for the "Custom APIs" tab in the menu bar or navigation panel. It should be the fifth option. * Click on the "Custom APIs" tab if necessary to access the APIs section.

2. **Selecting the API to Edit:** * Within the Custom API section, locate the API you want to edit from the list of available APIs. * Use the filtering options to narrow down the list of APIs based on specific criteria such as API name and function description. * Click on the API's name to access it's editing panel.

3. **Editing the API:** * **I**n the editing panel, make any appropriate changes needed. * After making the desired changes in the editing panel, review it for accuracy. * Once satisfied, click save and exit the editing panel to save the changes.

#### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage custom APIs. # Deleting a Custom API This section provides step-by-step instructions for deleting an API from your company. **Deleting a Custom API:** To remove an API from your company, follow these steps: 1. **Navigating to the Custom APIs Section:** * Log in to the admin panel using your credentials. * Look for the "Custom APIs" tab in the menu bar or navigation panel. It should be the fifth option. * Click on the "Custom APIs" tab if necessary to access the Custom API section.

2. **Selecting the Custom API to Delete:** * Within the Custom API section, locate the API you want to delete from the list of accessible APIs. * Click on the Custom API's name to access it's editing panel.

3. **Initiating the Deletion Process:** * Within the editing panel, look for a button at the bottom labeled "Delete API" or similar. * Click on the "Delete API" button to initiate the deletion process.

4. **Confirming API Deletion:** * A confirmation prompt may appear to ensure you intend to delete the API. * Review the confirmation message carefully. 5. **Finalizing Custom API Deletion:** * Confirm the deletion action if prompted. * Once confirmed, the API will be permanently deleted from the system. **Note:** Deleting a Custom API is irreversible and will result in removing user's access to the deleted API information. Ensure that you intend to delete the API and review the action carefully before confirming. **Note:** All interactions within the platform are recorded for compliance purposes. These records will remain in the system even after an API is removed. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage Custom APIs. # Custom Database The custom database is designed to meet the unique needs and requirements. Unlike storing data in a standard way it stores data, manages and processes data in a particular way. **Accessing the Custom Database:** To access your custom database in our application, follow these steps: 1. **Navigating to the Custom Database Section:** * Log in to the admin panel using your credentials. * Look for the "Custom Database" tab in the menu bar or navigation panel. It should be the sixth section you see when opening the admin panel.

2. **Adding a new Custom Database:** * To add a new Custom Database, click the "add new database" button.

3. **Filling out the Custom Database:** * A new window will open with option to fill the Name, SQL Server Connection String and Function Description. * *Name:* here you specify the database’s name. * *SQL Server Connection String:* connects to database server. When you add the string one more option will come which is table name. * *Function Description*: explains the database’s role. * *Table*: represents a specific data structure inside a database. * StockData - stores historical stock prices * StockDetails - stores stock details * StockView1 -for everything in one place. * Table Schema: explain the information the table holds.

4. **To view and edit the existing Custom Database:** * Click the desired row you would like to view or edit.

# Users This section provides step-by-step instructions for managing users of the platform within your company. **Accessing Users:** To access your company's users in our application, follow these steps: 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

2. **Viewing Users:** * Upon accessing the users section, you will be presented with a list of users registered in the system. * The list may include user details such as email addresses and first and last names.

3. **Filtering Users:** * Use the search option to narrow down the list of users based on specific criteria such as email address, first name, or last name.

4. **Viewing User Details:** * Click on the user to view detailed information in the user panel. * This includes personal details, company roles, and assigned groups and agents.

**Note:** Ensure that you have the necessary permissions to access and manage user accounts. Exercise caution when performing actions that affect user accounts to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users. # Inviting a New User This section provides step-by-step instructions for inviting a new user to the platform within your company. **Inviting a New User:** To invite a new user to the platform, follow these steps: 1. **Navigate to the Home Screen:** * Access the application and log in with your credentials. * Once logged in, you will land on the Home Screen.

2. **Access the Admin Panel:** * On the Home Screen, locate the dropdown menu in the top-left corner of the interface. * Click on the dropdown menu to view its options.

3. **Navigate to the Users Section:** * Within the dropdown menu, locate and select the "Admin" option. This action will redirect you to the Admin Panel.

4. **Initiating the Invitation Process:** * Once you're in the Admin Panel, navigate to the "Manage Users" tab. * Click on the "Invite New User" button to initiate the invitation process.

5. **Completing the Invitation Form:** * Upon clicking the "Invite New User" button, an "Invite New User" panel will open. * In the invitation panel, enter the new user's details: * First name * Last name * Email address * Optional phone number * Select the user's default agent from the dropdown menu.

6. **Sending the Invitation:** * After entering the required information, review it for accuracy. * Once satisfied, click on the "Add User" button or similar action to send the invitation to the new user.

7. **Confirmation:** * Upon sending the invitation, a confirmation message may appear to indicate that the invitation has been sent successfully. * The new user will receive an email containing instructions on how to complete their registration and set up their account. **Note:** Ensure that you enter the new user's information accurately, including a valid email address, to ensure successful delivery of the invitation. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users. * **User Licenses:** Make sure your company has purchased enough licenses to accommodate the new user. If your company does not have enough licenses you will not be able to add a new user. # Deleting a User This section provides step-by-step instructions for deleting a user from your company. **Deleting a User:** To remove a user from your company, follow these steps: 1. **Accessing the Admin Panel:** * Log in to the admin panel using your credentials.

2. **Navigating to Users:** * Once logged in, locate and click on the "Manage Users" tab in the menu bar or navigation panel.

3. **Selecting the User to Delete:** * Within the "Manage Users" section, locate the user you want to delete from the list of registered users. * Click on the user's name or username to access their user panel.

4. **Initiating the Deletion Process:** * Within the user panel, look for a button at the bottom labeled "Delete User" or similar. * Click on the "Delete User" button to initiate the deletion process.

5. **Confirming User Deletion:** * A confirmation prompt may appear to ensure you intend to delete the user. * Review the confirmation message carefully. 6. **Finalizing User Deletion:** * Confirm the deletion action if prompted. * Once confirmed, the user will be permanently deleted from the system. **Note:** Deleting a user's account is irreversible and will result in removing the user's access to the platform. Ensure that you intend to delete the user and review the action carefully before confirming. **Note:** All interactions within the platform are recorded for compliance purposes. These records will remain in the system even after a user is removed. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users. # Roles In our system, there are four distinct user roles, each with specific permissions and responsibilities. Below is an overview of each user role: 1. **Company Admin:** * The Company Admin role is the highest level of authority within the system. * Responsibilities: * Manage all aspects of the system, including user accounts, groups, agents, and settings. * Configure system-wide settings and permissions. * Access all features and functionalities of the admin panel. * Permissions: * Full control over user management, including creating, deleting, and modifying user accounts. * Access to all system settings and configurations. * Ability to assign roles and permissions to other users. * Complete access to all features and functionalities of the system. 2. **User:** * The User role represents standard users of the system who perform day-to-day tasks. * Responsibilities: * Perform assigned tasks within the system based on their role and permissions. * Access and utilize features and functionalities assigned to them by the Company Admin. * Permissions: * Limited access to system settings and configurations. * Ability to perform tasks assigned to them by the Company Admin. * Access to features and functionalities relevant to their role. 3. **Monitor:** * The Monitor role is responsible for monitoring activities and generating reports within the system. * Responsibilities: * Monitor user activities, interactions, and system performance. * Generate reports and analytics based on collected data. * Permissions: * Access to monitoring tools and reporting functionalities. * Ability to view and analyze user activities and system performance metrics. * Limited access to system settings and configurations. 4. **Compliance:** * The Compliance role ensures adherence to regulatory requirements and internal policies within the system. * Responsibilities: * Enforce compliance standards and regulations within the organization. * Audit user activities and ensure compliance with company policies. * Permissions: * Access to compliance-related tools and functionalities. * Ability to perform audits and enforce compliance standards. * Limited access to system settings and configurations. **Note:** User roles can be customized and tailored to meet the specific needs and requirements of your organization. Additionally, users can be assigned multiple roles or custom roles based on their responsibilities within the system. # Accessing User Roles This section provides step-by-step instructions for accessing and managing user roles within your company. **Accessing User Roles:** To access user roles in our application, follow these steps: 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

3. **Filtering Users:** * Use the search option to narrow down the list of users based on specific criteria such as email address, first name, or last name. 4. **Viewing User Details:** * Click on the user to view detailed information in the user panel. * Locate the 'Roles' section and use the checkboxes to assign user roles.

**Note:** Ensure that you have the necessary permissions to access and manage user roles. Exercise caution when performing actions that affect user accounts to avoid unintended consequences. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users. # Assigning an Agent This section provides step-by-step instructions for assigning an agent to a user within your company. **Assigning an Agent:** To assign an agent to a user in our application, follow these steps: 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

2. **Viewing Users:** * Upon accessing the "Manage User's" section, you will be presented with a list of users registered in the system. * The list may include user details such as email addresses and first and last names. * Use the filtering options to narrow down the list of users based on specific criteria such as email address, first name, or last name.

3. **Accessing the User Panel:** * Click on the user's name or username to access their user panel.

4. **Adding a New Agent:** * In the user panel, navigate to the "Agents" section. * This section displays the agents currently assigned to the user.

5. **Selecting the Agent:** * Look for a dropdown menu labeled "Add Agent" within the Agents section. * Click on the "Add Agent" dropdown menu to select the agent being assigned to the user. * Choose the desired agent from the dropdown menu or search results.

6. **Saving Changes:** * Once you have selected the agent, ensure your selection has been correctly added to the list of agents available to the user. * Click save and exit the user panel to save the changes.

**Note:** Assigning a new agent to a user grants them access to resources or permissions associated with that agent. Ensure that you assign agents to users based on their roles and responsibilities within the organization. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users and agent assignment. # Setting a Default Agent This section provides step-by-step instructions for setting a default agent for a user within your company. **Setting a Default Agent:** To set a default agent for a user in our application, follow these steps: 1. **Navigating to the Users Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Users" tab in the menu bar or navigation panel. * Click on the "Manage Users" tab to access the users' section.

2. **Viewing Users:** * Upon accessing the manage user's section, you will be presented with a list of users registered in the system. * The list may include user details such as email addresses and first and last names. * Use the filtering options to narrow down the list of users based on specific criteria such as email address, first name, or last name.

3. **Accessing the User Panel:** * Click on the user's name or username to access their user panel.

4. **Adding a Default Agent:** * In the user panel, navigate to the "Agents" section. * This section displays the agents currently assigned to the user.

5. **Selecting the Agent:** * Look for a dropdown menu labeled "Default Agent" within the Agents section. * Click on the "Default Agent" dropdown menu to select the default agent being assigned to the user. * Choose the desired agent from the dropdown menu or search results. 6. **Saving Changes:** * Once you have selected the default agent, ensure your selection is correctly displayed in the 'Default Agent' dropdown menu. * Click save and exit the user panel to save the changes. **Note:** Setting a default agent for a user grants them immediate access to that agent upon signing in to the platform. Ensure that you assign agents to users based on their roles and responsibilities within the organization. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage users and agent assignment. # Groups This section provides step-by-step instructions for managing groups within the admin panel. **Accessing Groups:** To access groups in our application, follow these steps: 1. **Navigating to the Groups Section:** * Log in to the admin panel using your credentials. * Look for the "Manage Groups" tab in the menu bar or navigation panel. * Click on the "Manage Groups" tab to access the groups section.

2. **Viewing Groups:** * Upon accessing the groups section, you will be presented with a list of groups registered in the system. * The list may include group details such as its name, description, and user count.

3. **Filtering Groups:** * Use the search option to narrow down the list of groups if necessary. 4. **Viewing Group Details:** * Click on the group to view detailed information in the group editing panel. * These details include group name, description, and agents and users assigned to the group.

#### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Adding a New Group This section provides step-by-step instructions for adding a new group within the admin panel, including naming, describing, adding agents, and assigning users to the group. **Adding a New Group:** To add a new group in our application, follow these steps: 1. **Accessing the Groups Section:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the "Add New Group" button. This action triggers the creation process.

2. **Configuring Group Details:** * Upon clicking "Add New Group," a "Create New Group" panel will appear. * Provide a suitable name for the new group in the designated field. * Add a brief description to help identify the group's purpose or function. * If you have the GroupMember.Read.All Permission you can assign an entra id group * ```

``` 3. **Adding Agents:** * To include agents in the group, locate the "Add Agents" section within the panel. * Select the agents you wish to associate with the group from the available list. * Once selected, confirm the addition by clicking on the appropriate agent.

4. **Assigning Users:** * In the "Add Users" section, specify the users who will have access to the group. * Once selected, confirm the addition by clicking on the appropriate user.

5. **Finalizing Group Creation:** * Review all the details entered for accuracy. * Once satisfied, proceed to finalize the creation of the group by clicking save on the 'Create New Group' panel. * The new group will now be available for use with the assigned agents and users.

**Note:** Ensure that the group name, description, assigned agents, and users accurately reflect the intended configuration before finalizing the creation process. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Deleting a Group This section outlines the steps to delete a group within the admin panel, starting from accessing the group tab to confirming the deletion in the group edit panel. **Deleting a Group:** To delete a group in our application, follow these steps: 1. **Accessing the Group Edit Panel:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the group from in which you want to delete. This action opens the group edit panel.

2. **Navigating to Group Edit Panel:** * Upon clicking the group name, you will be directed to the group edit panel. * This panel displays the details and settings associated with the selected group.

3. **Initiating Deletion Process:** * Scroll down to the bottom of the group edit panel. * Locate the "Delete Group" option or button.

4. **Confirming Deletion:** * Click on the "Delete Group" button. A confirmation dialog may appear to ensure you intend to proceed with the deletion. * Confirm the deletion action if prompted. 5. **Finalizing Deletion:** * After confirming, the group will be scheduled for deletion. * Once deleted, the group and all associated settings will be permanently removed from the system. **Note:** Ensure that you have selected the correct group for deletion, as this action is irreversible and may impact associated agents and users' access permissions. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Assigning a User This section outlines the steps to assign a user to an existing group within the admin panel, from accessing the group tab to confirming the user assignment in the group details. **Assigning a User to a Group:** To assign a user to a group in our application, follow these steps: 1. **Accessing the Group Edit Panel:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the group from which you want to assign a user. This action opens the group edit panel.

2. **Navigating to the Users Section:** * Once in the group edit panel, scroll down to find the "Users" section. * This section displays the users currently assigned to the group.

3. **Adding a User:** * In the "Users" section, you will find an option to add a user to the group. * Click on the dropdown menu provided to select the user you want to add.

4. **Selecting the User:** * A dropdown menu will appear containing a list of available users. * Scroll through the list or use the search functionality to find the user you wish to add. * Click on the user's name to select them.

5. **Confirming User Assignment:** * After selecting the user, ensure that the correct user is displayed in the list of users within the 'Users' section. * Review the selection to verify accuracy. 6. **Saving Changes:** * Once you have ensured they have been added to the group, click save and exit the group editing panel to save the changes.

**Note:** Ensure that you assign users to groups based on their roles and responsibilities within the organization to maintain appropriate access levels and permissions. #### Bulk adding users in EntraID If you have assigned an entra id security group you can also bulk load users by going to the group in entra id and using the Bulk upload function seen below for more information visit

#### #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Removing a User This section outlines the steps to remove a user from an existing group within the admin panel. **Removing a User from a Group:** To remove a user from a group in our application, follow these steps: 1. **Accessing the Group Edit Panel:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the group from which you want to remove a user. This action opens the group edit panel.

2. **Navigating to the Users Section:** * Scroll down within the group edit panel to find the "Users" section. * This section displays the users currently assigned to the group.

3. **Identifying the User to Remove:** * In the "Users" section, locate the user you wish to remove from the group.

4. **Initiating User Removal:** * Next to the user's name, there should be an option or icon indicating removal or deletion. * Click on this option to initiate the removal process.

5. **Confirming User Removal:** * A confirmation prompt may appear to ensure you intend to remove the user from the group. * Review the confirmation message carefully. 6. **Finalizing User Removal:** * Confirm the removal action if prompted. * Once confirmed, the user will be removed from the group, revoking their access to any resources or permissions associated with that group. **Note:** Removing a user from a group does not delete the user's account or profile from the system. It only removes their association with the specific group. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Adding an Agent This section provides step-by-step instructions for adding an agent to a group within the admin panel. **Adding an Agent to a Group:** To add an agent to a group in our application, follow these steps: 1. **Accessing the Group Edit Panel:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the group to which you want to add an agent. This action opens the group edit panel.

2. **Navigating to the Agents Section:** * Scroll down within the group edit panel to find the "Agents" section. * This section displays the agents currently assigned to the group.

3. **Identifying the Agent to Add:** * In the "Agents" section, locate the agent you wish to add to the group in the dropdown menu.

4. **Initiating Agent Addition:** * In the "Agents" section, you will find an option to add an agent to the group. * Click on the dropdown menu provided to select the agent you want to add.

5. **Confirming Agent Addition:** * After selecting the agent, ensure that the correct agent is displayed in the list of agents within the 'Agents' section. * Review the selection to verify accuracy. 6. **Finalizing Agent Addition:** * Once you have ensured the agent has been added to the group, click save and exit the group editing panel to save the changes. * The agent is now added to the group.

**Note:** Adding an agent to a group allows the group's users to access it. Ensure that you assign agents to groups based on their roles and responsibilities within the organization. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Removing an Agent This section provides step-by-step instructions for removing an agent from a group within the admin panel. **Removing an Agent from a Group:** To remove an agent from a group in our application, follow these steps: 1. **Accessing the Group Edit Panel:** * Log in to the admin panel using your credentials. * Navigate to the "Manage Groups" tab in the menu bar. * Click on the group from which you want to remove an agent. This action opens the group edit panel.

2. **Navigating to the Agents Section:** * Scroll down within the group edit panel to locate the "Agents" section. * This section displays the agents currently assigned to the group.

3. **Identifying the Agent to Remove:** * In the "Agents" section, identify the agent you wish to remove from the group.

4. **Initiating Agent Removal:** * Next to the agent's name, there should be an option or icon indicating removal or deletion. * Click on this option to initiate the removal process.

5. **Confirming Agent Removal:** * A confirmation prompt may appear to ensure you intend to remove the agent from the group. * Review the confirmation message carefully. 6. **Finalizing Agent Removal:** * Confirm the removal action if prompted. * Once confirmed, click save and exit the group editing panel to save the changes.

**Note:** Removing an agent from a group removes the group's user's ability to use that agent. Ensure that you intend to remove the agent and review the action carefully before confirming. #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage groups. # Agent Scanner Defaults You can set default values for each new agent created using scanner defaults. There are seven distinct scanners available, and you have the option to enable or disable each one independently. **Accessing Agent Scanner Defaults:** To access the agent scanner defaults in our application, follow these steps: 1. **Navigating to the Agent Scanner Defaults Section:** * Log in to the admin panel using your credentials. * Look for the "Scanner Defaults" tab in the menu bar or navigation panel. * Click on the "Scanner Defaults" tab to access the Agent Scanner Defaults section.

2. **Viewing Agent Scanner Defaults:** * Upon accessing the agent scanner defaults section, you will be presented with a list of available options in the system.

To watch a video demonstration, click [here](https://airiskinc-my.sharepoint.com/:v:/g/personal/vetrimaheswaran_aicrisk_com/EcVSQVnGcolLkX7If2uqNpQBCula263TSTBOLFlEcuqfbw?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJPbmVEcml2ZUZvckJ1c2luZXNzIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXciLCJyZWZlcnJhbFZpZXciOiJNeUZpbGVzTGlua0NvcHkifX0\&e=EcFOTr). #### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and manage the agent scanner defaults. # Toxicity Scan for toxic language:

**Purpose:** The Toxicity Scanner is designed to identify and flag content that exhibits toxic or harmful behavior, such as hate speech, threats, or harassment. **Functionality:** It employs natural language processing (NLP) techniques to analyze text inputs and determine the presence of toxic language. This includes assessing the tone, context, and semantic meaning of the text to identify potentially harmful content. **Implementation:** The scanner utilizes a combination of machine learning algorithms and pre-trained models to recognize patterns indicative of toxicity. It may employ techniques such as sentiment analysis, keyword matching, and context analysis to accurately detect toxic behavior. **Usage:** The Toxicity Scanner can be integrated into agents to automatically moderate user-generated content and maintain a safe and respectful environment. To learn more about toxicity and other LLM parameters visit: {% embed url="" %} {% endembed %} # Personal Information Scan for sensitive personal information:

**Purpose:** The Personal Information Scanner is designed to detect and redact sensitive personal information (PII) from text inputs to ensure user privacy and data protection. **Functionality:** It utilizes pattern recognition algorithms and regular expressions to identify common types of PII, such as names, addresses, phone numbers, email addresses, and social security numbers, within text data. **Implementation:** The scanner employs a set of predefined rules and patterns to identify and extract PII from text inputs. This may include matching specific formats, such as email addresses or phone numbers, as well as analyzing contextual clues to accurately identify personal information. **Usage:** The Personal Information Scanner is commonly used in business environments to automatically detect and remove sensitive information before it is shared or stored. # Topic Scanner Scan for inappropriate and unwanted topics:

**Purpose:** The Topic Scanner is designed to analyze text inputs and identify the main topics or themes discussed within the content. **Functionality:** It employs topic modeling techniques to extract key topics from text data based on the distribution of words and their relationships. **Implementation:** The scanner processes text inputs using probabilistic algorithms to identify clusters of words that frequently co-occur and represent coherent topics. It may also incorporate natural language understanding (NLU) techniques to infer the meaning and context of the text. **Usage:** The Topic Scanner is utilized to understand and categorize textual content automatically. # Prompt Injection Detection Scan for misuse and attempts at manipulating an agent:

**Purpose:** The Prompt Injection Detection scanner aims to identify instances where malicious prompts or commands are injected into text inputs to manipulate the system into doing something it should not. **Functionality:** It analyzes text inputs for signs of unexpected or unauthorized commands or instructions that could potentially compromise the security or integrity of the system. **Implementation:** The scanner employs a combination of pattern matching, anomaly detection, and contextual analysis techniques to identify suspicious commands or prompts within text data. It may also leverage machine learning models trained on known examples of prompt injections to improve detection accuracy. **Usage:** Prompt Injection Detection is commonly used in security-critical applications, such as command-line interfaces, chatbots, and voice assistants, to prevent unauthorized access or manipulation of systems. To learn more about prompt injections visit: {% embed url="" %} {% endembed %} To learn more about temperature visit: {% embed url="" %} {% endembed %} # Regex Scanner Scan for custom text and strings:

**Purpose:** The Regex Scanner is designed to search for and extract text patterns that match specific regular expressions (regex) within text data. **Functionality:** It applies user-defined or predefined regex patterns to text inputs to identify and extract information that conforms to those patterns. **Implementation:** The scanner processes text inputs using efficient pattern matching algorithms to locate substrings that match the specified regex patterns. It supports a wide range of regex syntax and features for flexible pattern matching. **Usage:** Regex Scanner is commonly used in text processing tasks such as data extraction, text validation, and text transformation, where precise pattern matching is required. To create a regex for a custom string, try this intuitive regex builder: {% embed url="" %} {% endembed %} # Geographic Gating Restrict agent use to a geographic location:

**Purpose:** The Geographic Gating scanner is used to restrict access to an agent based on the user's geographical location. **Functionality:** It analyzes network traffic or user metadata to determine the geographical location of the platform user. **Implementation:** The scanner employs geolocation databases or services to map IP addresses or other location indicators to geographic coordinates. It then compares this information against predefined access policies or geographic boundaries to enforce access controls or filter content. **Usage:** Geographic Gating is commonly used in content delivery networks (CDNs), streaming services, and online platforms to enforce regional licensing agreements, comply with legal requirements, or customize content based on user location. # Language Detection Detect foreign languages and syntax used in agent prompts:

**Purpose:** The Language Detection scanner is designed to automatically identify the language(s) used in text inputs. **Functionality:** It analyzes the linguistic features of text data, such as vocabulary, syntax, and character n-grams, to determine the most likely language(s) present in the input. **Implementation:** The scanner employs statistical language models or machine learning algorithms trained on multilingual text corpora to recognize patterns indicative of different languages. It may use techniques such as n-gram analysis, language-specific word lists, or deep learning models to improve accuracy. **Usage:** Language Detection is commonly used in multilingual applications, text processing pipelines, and internationalization efforts to automatically adapt language-specific processing or content presentation based on user input. # Allowed File Types Restrict attachments used with agents to specific file types:

**Purpose:** The Allowed File Types scanner is designed to automatically identify unwanted files used in file uploads and attachments. **Functionality:** It analyzes the attached file type, such as .txt, .xml, and .pdf, to determine if it is allowed to work with the agent. **Usage:** The Allowed File Type selector is commonly used with agents that are commonly analyzing data from a static file. It allows the agent to serve a very specific purpose within your company. Please make sure that your files are in one of the supported formats before uploading them to an agent:

# Logs Logs are a vital aspect of monitoring and tracking system activities within the application. They provide a record of events, errors, and changes made within the admin panel. **Accessing Logs:** To access the logs in our application, follow these steps: 1. **Navigating to the Logs Section:** * Log in to the admin panel using your credentials. * Look for the "Logs" tab in the menu bar or navigation panel. * Click on the "Logs" tab to access the Logs section.

2. **Viewing Logs:** * Upon accessing the logs section, you will be presented with a list of logs registered in the system. * The list may include log details such as the date created, user ID, and a message.

3. **Filtering Logs:** * Use the search option to narrow down the list of logs if necessary.

#### Additional Information: * **Company Admin Privileges**: Please note that only users with company admin privileges can access the admin panel and logs. # Interpreting Logs Once you've accessed the logs, you will see a list of entries containing information such as timestamps, user actions, and system events. **Interpreting Logs:**

Each log entry typically includes the following details: * **Timestamp:** Indicates when the event occurred.

* **User:** Identifies the user who initiated the action or event.

* **Action/Event Message:** Describes the action or event that took place.

**Common Use Cases for Viewing Logs:** * **Troubleshooting:** Identify errors or issues within the system and trace their origins. * **Auditing:** Track user actions and system events for compliance and accountability purposes. * **Security Monitoring:** Detect and investigate suspicious activities or unauthorized access attempts. * **Performance Monitoring:** Analyze system performance and identify areas for optimization. # Server Admin # Upgrade This section provides step-by-step instructions for navigating the Upgrade section. **Navigating to the Upgrade Section:** * Log in to the server admin panel using your credentials. * Look for the "Upgrade" tab in the menu bar or navigation panel. * Click on the "Upgrade" tab to access the upgrade section.

**Server Version:**

**Release Date:**

**Release Notes:**

**Action:** * In the action panel, make any appropriate changes needed. * After making the desired changes in the action panel, review it for accuracy. * Once satisfied, click save and exit the editing panel to save the changes.

# Default Agents This section provides step-by-step instructions for navigating the Default Agents section. **Navigating to the Default Agents Section:** * Log in to the server admin panel using your credentials. * Look for the "Default Agent" tab in the menu bar or navigation panel. * Click on the "Default Agent" tab to access the default agent section.

**BOT Name:**

**BOT Description:**

**BOT Type:**

# Adding a New Agent This section provides step-by-step instructions for adding a new agent. 1. **Navigating to the Default Agents Section:** * Log in to the server admin panel using your credentials. * Look for the "Default Agent" tab in the menu bar or navigation panel. * Click on the "Default Agent" tab to access the default agent section. 2. **Accessing the Add New Agent Panel:** * Within the "Default Agents" section, locate the button labeled "Add New Agent." * Click on the "Add New Agent" button to access the Create New Agent panel.

3. **Creating the New Agent:** * Within the Create New Agent panel, enter the new agent details. There are three sections to review. * Default Bot Preview

* General

* GPT Settings

4. **Finalizing the New Agent:** * After entering the required information, review it for accuracy. * Once satisfied, click save and exit the create new agent panel to save the changes.

# Data Archive This section provides step-by-step instructions for navigating the Data Archive section. **Navigating to the Data Archive Section:** * Log in to the server admin panel using your credentials. * Look for the "Data Archive" tab in the menu bar or navigation panel. * Click on the "Data Archive" tab to access the data archive section.

**Conversation:**

**Compliance:**

**Files:**

**Time:**

# API Access This section provides step-by-step instructions for navigating the API Access section. **Navigating to the API Access Section:** * Log in to the server admin panel using your credentials. * Look for the "API Access" tab in the menu bar or navigation panel. * Click on the "API Access" tab to access the API Access section.

**Description:**

**Client ID:**

**Redirect URLS:**

**Action:** * In the action panel, make any appropriate changes needed. * After making the desired changes in the action panel, review it for accuracy.

**Adding a new API user:** To add a new API User to your company, follow these steps: 1. **Navigating to the API Users Section:** * Log in to the server admin panel using your credentials. * Look for the "API Access" tab in the menu bar or navigation panel. It should be the fourth option. * Click on the "API Access" tab if necessary to access the API Access section. 2. **Accessing the Add New API User Panel:** * Within the API Access section, locate the button labeled "Add API User". * Click on the "Add API User" button to access the Add API user panel.

3. **Adding the New API User:** * Within the Add API User panel, enter the new user details. There are three sections to review. * Description:

* Client ID:

* Redirect URLS:

4. **Finalizing the New API User:** * After entering the required information, review it for accuracy. * Once satisfied, click save and exit the create new API user panel to save the changes.

# Chat # List Agents List all agents the user has access to.

**Type:** GET **End Point:** /api/Chat/ListAgents **Parameters:** No parameters # Create Conversation Creates a conversation instance of a specific user and agent.

**Type:** GET **End Point:** /api/Chat/CreateConversation/{agentid} **Parameters:** * **agentid-string (path)** * *Description: agentid* * **title-string (query)** * *Description: title* * **MetaData-array\[string] (query)** --- [Next Page](/llms-full.txt/1)