Skip to main content

​
Instance Management

Instances are the core of your Praxis AI experience, allowing you to manage and interact with various digital twins.
This feature is only available for Digital Twin Instance Admins
Admin Configurations Main Ui Gi Start by clicking the Admin dashboard, located at the top of the main user interface. The Instances tab displays all instances your user is entitled to manage and also allows you to create a new instance if you are entitled to do so. Configurations Instances Tab List Web

​
Viewing Instances

List of Instances

The Instances tab displays all instances you have access to. Each instance is listed with its name, status, and other relevant details.

Instance Details

Clicking on an instance will provide you with detailed information, including its configuration, members, and activity logs.
Depending on your administrative rights, your editing privileges may be limited.
Instances General Settings The Configuration section allows you to manage your instance settings. Configure these key features to optimize your digital twin’s performance and accessibility.
Set your instance name using consistent naming conventions:
Canvas integration: instance_name.school_name.instructure.com
Praxis LXP integration: instance_name.praxislxp.com
The credit balance for this instance. New instances receive 50 starter credits. Purchase additional credits at https://pria.praxislxp.com/my-profile/pricing or contact your institution’s IT administrator for enterprise options.
To adjust this field, contact the Praxis AI team at humans@praxis-ai.com.
Controls instance availability:
  • Active — Instance is operational for all users
  • Inactive — Instance is temporarily disabled
To adjust this field, contact the Praxis AI team at humans@praxis-ai.com.
Unique identifier automatically generated during instance creation. Used for API integrations, Chat Completions, and system connections. This field is read-only — use the copy button to copy it to your clipboard.
Specify administrator email addresses (comma-separated) for instance monitoring and notifications. Used for moderation alerts and system notifications.
Links your instance to an institutional account for centralized credit management and distribution across multiple digital twins. Select from available accounts in the dropdown.
Controls how new users can join this instance:
  • Disabled — Users cannot self-join; they must be added by an administrator
  • Account only — Only users within the same institutional account can join
  • Everyone (public) — Anyone can join this instance
To adjust this field, contact the Praxis AI team at humans@praxis-ai.com.
When enabled, only administrators can join the instance — regular users are blocked from self-joining. This option appears only when Allow Joining is set to “Account only” or “Everyone (public)”.
To adjust this field, contact the Praxis AI team at humans@praxis-ai.com.
Enable credit sharing across all instance users. Each interaction consumes credits based on complexity. When credits are depleted, users cannot interact until credits are replenished according to your service agreement.
When Pool Credits is disabled, requests from users draw from each user’s personal credit pool instead.
The number of credits automatically granted to new users when they first join this instance. Set to 0 to disable automatic credit awards. This field is only visible when Pool Credits is disabled (i.e., users manage their own credit balance).
Automatically grants full entitlements to new administrators, providing complete dashboard access and customization capabilities. For specific entitlement questions, contact humans@praxis-ai.com.
Screens every user message for unsafe content before it reaches the AI model. Flagged messages are blocked with a safe response, and an alert email goes to the instance’s Contact Emails.
FieldDescriptionDefault
Enable ModerationTurns content screening on or offOn
Moderation ModelThe moderation model to use (e.g., omni-moderation-latest)omni-moderation-latest
For the category taxonomy, model options, what users see when a message is blocked, and how to review flagged conversations, see Content Moderation.
When enabled, the Digital Twin always includes source citations in its responses. This instructs the AI to reference the documents, files, or knowledge base entries it used to generate each answer, improving transparency and traceability.
Disables the initial personalization popup for new users, allowing direct access to the conversation interface without being prompted to set preferences first.
Enables location-based responses for queries about local services, weather, campus information, and regional content. Improves response accuracy and relevance for location-dependent requests.
Activates real-time voice communication capabilities. Users can speak directly to the Digital Twin and receive verbal responses in real-time. See Convo Mode for details.When enabled, additional Convo settings appear below.
Restricts Convo Mode to administrators only — regular users will not see the Convo button. Visible only when Enable Convo is turned on.
Allows users to type text messages during a Convo session in addition to speaking. Visible only when Enable Convo is turned on.
Voice and audio settings (Speech-to-Speech Model, Voice Selection, Voice Activity Detection, Noise Reduction, Transcription Language, and ElevenLabs Connection Method) are now configured in the Personalization tab under the Convo section. See Personalization for details.
If ElevenLabs is selected as the voice provider, the ElevenLabs Agent ID and API Key must still be configured in the Integrations section on this page.
Shows detailed information about the tool calls and agent actions made to respond to queries. When enabled, users can see which tools were invoked and their results in the conversation interface.
When enabled, users are locked to this instance and cannot navigate to or switch to other Digital Twin instances. Useful for kiosk-style deployments or single-purpose integrations.
Prevents non-admin users from copying AI-generated content from conversations. When enabled:
  • Text selection is disabled on AI responses and user inputs
  • Copy keyboard shortcuts (Ctrl+C / Cmd+C) are blocked
  • Copy-to-clipboard buttons are hidden
  • Share link buttons are hidden
Administrators retain full clipboard access regardless of this setting.
Useful in exam or assessment environments, intellectual property protection scenarios, or when AI-generated content must not be easily extracted.
Hides the option for users to purchase or add credits to their account. Use this when credit management is handled exclusively by administrators or through institutional agreements.
Controls automatic conversation history management to keep conversations within AI model context limits.When enabled, Pria uses LLM summarization to compress older messages in long conversations. This allows conversations to continue indefinitely without losing context — earlier messages are summarized rather than truncated.
FieldDescriptionDefault
Compact HistoryEnable automatic history compactionOn
The K-Mean Score field nearby is a retrieval setting, not a compaction one — it sets the relevance floor for knowledge search. See Knowledge & RAG Configuration.
How it works:
  1. As a conversation grows, Pria monitors the total token count
  2. When approaching the model’s context limit, older messages are grouped and summarized by an LLM
  3. The summaries replace the original messages, preserving key context while reducing token usage
  4. Base64-encoded data (images, files) is automatically stripped from compacted history
  5. Tool results are enriched with human-readable summaries before compaction
History compaction is recommended for instances where users have long, ongoing conversations (e.g., research projects, multi-session tutoring). For short Q&A interactions, it’s unnecessary.

​
Knowledge & RAG Settings

These settings control how your Digital Twin retrieves information from the IP Vault (uploaded files, links, audio notes). The full reference — retrieval mode defaults (Disabled / Normal / KAG Fusion / Search Only), KAG enablement, chunk handling, the kmeanScore similarity threshold, embedding model selection, and file upload policies (disableFileUploadForUser, enableFileUploadForEmail, maxFiles, ignorePersonalVault) — lives on the dedicated Knowledge & RAG Configuration page. On this Configuration tab you will also find the display toggles that decide what users see alongside each response:
  • Display RAG Search Details — show which passages were retrieved and their relevance scores
  • Display Thinking Details — show the model’s reasoning trace when thinking is enabled
  • Display Tool Execution — show real-time progress as tools execute
  • Display Agent Details — show the complete agent step-by-step breakdown
These four toggles are independent and default to on. Turn them off for kiosk-style deployments where the conversation should feel chat-like rather than tool-driven.

​
Memory Settings

The Digital Twin remembers facts about each user across conversations (memory parameters), and long conversations are automatically summarized to stay within the model’s context window (Compact History — see the History Compaction accordion above). For the memory model — personal vs. instance-shared memory, how rows are written during conversations, and how admins inspect or edit them — see Memory Parameters & Act As.

​
Tool & Assistant Controls

Govern which tools and assistants are available to standard users on this instance.
Field: toolsDisabled Default: NoneA multi-select of platform tools that the Digital Twin should NOT use on this instance — e.g., disabling generate_image for a text-only academic instance, or hiding call_canvas when Canvas is not configured. Tools are matched by ID and excluded from the model’s tool list at request time, so even if the model “tries” to call them, the request is rejected upstream.
Field: toolResultsMaxChars Default: 60000Maximum number of characters returned by a single tool call. Longer results are truncated. Increase for instances that work with large documents or detailed API responses. Decrease to reduce token usage and protect against runaway tool output.
Field: disableAssistantsForUser Default: OffWhen enabled, regular users cannot access the Assistant Library or use custom assistants. Administrators retain full access. Useful when the Digital Twin should present a single unified persona rather than a menu of specialists.
Field: disableCreateAssistantsForUser Default: OffWhen enabled, regular users can use assistants but cannot create new ones. All assistants are admin-authored. Useful for curated deployments (curriculum, customer support).
Field: assistantsDisabled Default: NoneA multi-select of specific assistants that should be hidden from the assistant library on this instance, even if they exist in the catalog. Use this to retire assistants without deleting them.
Field: enableAssistantsForEmail Default: EmptyWhitelist of email addresses that are allowed to use assistants when assistants are otherwise disabled for users. Useful for granting authoring access to a small pilot group.

​
Realtime Voice (Convo Mode)

The high-level voice toggles — Enable Convo (Speech to Speech) and Convo Admin Only — are covered in the Instance Management accordions above. Provider choice (OpenAI GPT-Realtime, Gemini Live, ElevenLabs, Anam Avatar, xAI Voice), voice selection, voice activity detection, noise reduction, transcription language, and avatar customization all live in the Personalization tab — see Realtime Voice & Avatars for the full setup guide.

​
Provider API Keys (Bring Your Own Key)

By default, the Digital Twin uses Praxis-managed credits for every AI provider. When you provide your own API key for a provider, the Digital Twin routes that provider’s requests directly to your account — bypassing Praxis credits for that provider and giving you full control over billing, rate limits, and model availability. Keys can be supplied for OpenAI, Anthropic, Google Gemini, Mistral AI, xAI, Stability AI, Cohere, DeepSeek, OpenRouter, and InferX, plus the voice/avatar providers ElevenLabs, Anam, and LemonSlice (which also take an Agent/Avatar ID). The per-provider fields and key-generation links are in the API Keys section of the Integrations tab, below.
How BYOK routing works: When a model is selected for a Model Use (Personalization) and an API key is configured for that model’s provider, the request bypasses Praxis credits and routes directly with your key. If no key is set, the request falls back to Praxis credits. The fallback is per-provider, so you can BYOK for OpenAI while still using Praxis credits for Anthropic, for example.
Security: All API keys are stored encrypted, excluded from API responses by default, and never exposed to the browser. Only administrators with the appropriate entitlements can view or modify them.

​
MCP Servers

Configure external Model Context Protocol (MCP) servers that the Digital Twin can call as tools. MCP is the open standard for connecting LLMs to external systems — your Digital Twin can use MCP servers as additional tool sources alongside Pria’s built-in tools.
Field: mcpServerEnabled Default: OffMaster switch for outbound MCP connections from this Digital Twin. When enabled, configured MCP servers appear in the tool catalog and are invoked alongside built-in tools.
Field: mcpServerSecretUnique authentication token used when this Digital Twin exposes itself as an MCP endpoint (inbound MCP). Generated automatically; can be rotated. Configure third-party MCP clients with this secret to call your Digital Twin as an MCP source.
For the full MCP setup — including authoring per-server configurations, supplying credentials, and exposing your Digital Twin as an MCP source — see MCP Servers.

​
Advanced Configuration

These settings control fine-grained behavior of your instance. Most administrators can leave these at their default values.
FieldDescriptionDefault
Themedark or light — sets the default UI theme for the instancelight
Custom CSSInject custom CSS to style the Pria interface. Use this for branding (colors, fonts, logos)Empty
Custom CSS is injected as-is into the page. Test thoroughly to avoid breaking the layout.

​
Integrations

Admin Configurations Integrations Gi The Integrations section enables you to connect your digital twin instance with various platforms and services to extend functionality and streamline workflows.

​
Public Authorized URLs

Configure the URLs authorized to access your instance for secure integration with external platforms. Example for Canvas Integration: 
https://<your_school>.instructure.com
https://<your_school>.beta.instructure.com
https://canvas.instructure.com
All inbound requests are validated against this authorized URL list. Only requests originating from these URLs will be permitted to access your instance. For the hostname matching rules (subdomains, ports, what a broad entry authorizes), see Branding & Public URLs.

​
LTI Contexts

LTI (Learning Tools Interoperability) contexts are automatically populated when you connect an LTI placement to your digital twin instance. Key Features:
  • Automatic Population: Contexts are created when LTI placements are established
  • Instance Memory: The system remembers which instance to launch when accessing your digital twin through LTI or Canvas theme-based integration
  • Easy Management: Remove placements directly from this list or through your profile page
To disconnect an LTI integration:
  1. Remove the placement from the LTI Contexts list, or
  2. Navigate to your profile page for additional management options

​
Google Workspace

Enable Google Workspace services (Gmail, Drive, Calendar, Sheets, Docs, Slides, Meet, Classroom) for your Digital Twin from this section. You choose between two identity models: each user connects their own Google account, or you turn on Use Digital Twin Identity and connect one shared institutional account for everyone via the Connect Google Account OAuth flow. The full setup walkthrough — the service capability table, service dependencies, the two connection models, and the connect/disconnect flow — is on the Cloud Services page; see also the Google Workspace Integration guide for the authorization model and troubleshooting.
When institution-shared credentials are active, they take precedence over any personal Google credentials a user may have configured. If an admin connects a shared Gmail, all users will use that shared Gmail — even if they’ve also authorized their own personal account.

​
ElevenLabs

Connect your Digital Twin to an ElevenLabs Conversational AI agent to enable voice capabilities — both for Pria’s built-in Convo Mode and for embeddable client widgets. Two credentials are entered in this section:
  • ElevenLabs Agent ID — the Agent ID from your ElevenLabs Conversational AI agent
  • ElevenLabs API Key — authenticates requests between ElevenLabs and your Digital Twin; must match the key in your ElevenLabs agent’s Custom LLM settings
If no institution-level API key is set, the system falls back to the platform-wide ElevenLabs key (if configured).
Once the credentials are saved, select ElevenLabs as the Convo Mode provider under Personalization and AI Models — see Realtime Voice & Avatars. For the full end-to-end setup (creating the agent, configuring the Custom LLM connection, deploying client widgets), see the ElevenLabs Voice Agent integration guide.

​
Canvas

Connect your Digital Twin to Canvas LMS so it can query courses, assignments, grades, and more through the Canvas REST APIs. If these credentials are not configured, the call_canvas and search_canvas tools will be unavailable.
The numeric identifier from your Canvas Developer Key (e.g., 10000000000217). This is generated when a Canvas administrator creates an API-type Developer Key under Admin → Developer Keys.One Developer Key can be shared across multiple Digital Twin instances within the same institution.
The secret key paired with the Client ID above. Click Show Key on your Canvas Developer Key to reveal and copy it. Together with the Client ID, this enables OAuth2 authentication so users can authorize the Digital Twin to access Canvas on their behalf.
A personal access token generated by a registered course instructor. This is optional but required for teacher-level operations such as posting grade comments, creating announcements, or performing bulk grading — actions that require instructor-scoped permissions even when the end user is a student.To generate one: in Canvas, go to Account → Settings → New Access Token.
This token grants instructor-level access to the courses the faculty member is enrolled in. Store it securely and rotate it periodically.
A whitelist of Canvas REST API endpoints your Digital Twin is allowed to call. Each scope follows the syntax:
url:<HTTP_METHOD>|<API_ENDPOINT>
For example: url:GET|/api/v1/courses/:course_id/smartsearchScopes must be configured in two places — on the Canvas Developer Key and here in your Digital Twin instance. If a scope is missing from either location, the API call will be rejected.
During development, disable scope enforcement on your Canvas Developer Key to discover which endpoints your Digital Twin needs. Monitor the Agent Details in your dialog history to see which API calls are made, then add those scopes here for production. Use the Upload/Download buttons to manage the list from a JSON file.
See Canvas Scopes Reference for the full list of available scopes and a starter configuration.
When enabled, users must authenticate with Canvas and generate an authorization token before accessing the Digital Twin through the Web SDK or LTI integration. This adds an extra security layer by requiring a valid Canvas session upfront, rather than prompting for authorization only when a Canvas tool is first invoked.
For a complete walkthrough — including creating the Developer Key in Canvas, configuring scopes, testing your integration, and troubleshooting — see the Canvas Pria Tools integration guide.

​
Kaltura

Connect your Digital Twin to the Kaltura Video Platform to enable video search, playback, and media-related interactions within conversations. Enter your Kaltura Partner ID and API Secret (plus a Client ID when using a user-level secret) in this section — see Cloud Services for the field reference and where to find each value in the Kaltura Management Console.

​
Amazon Web Services (AWS)

Connect your Digital Twin to your own AWS services — primarily Amazon Bedrock for AI model access and Amazon S3 for storage. Administrators configure their own IAM credentials so the Digital Twin uses their AWS account directly, allowing access to Bedrock foundation models (e.g., Claude, Titan, Llama) and S3 storage under your organization’s billing and security policies.
The AWS region where your services are deployed. Select the region closest to your users or where your Bedrock models and S3 buckets are provisioned.Common choices:
  • us-east-1 (N. Virginia) — Broadest Bedrock model availability
  • us-west-2 (Oregon) — Good Bedrock coverage
  • eu-central-1 (Frankfurt) — European data residency
  • ap-southeast-1 (Singapore) — Asia-Pacific
Not all Bedrock foundation models are available in every region. Check the AWS Bedrock model availability page for your region.
The access key ID from your AWS IAM credentials. Generate this in the AWS IAM Console by creating an IAM user or role with the necessary permissions for Bedrock and/or S3.
The secret access key paired with the Access Key ID above. This is shown only once when you create the IAM credentials in AWS — store it securely.
This is a sensitive credential. It is stored encrypted and excluded from API responses by default. Never share it in client-side code or logs.
IAM Permissions: Your IAM user or role needs the following policies:
  • Amazon Bedrock — bedrock:InvokeModel, bedrock:InvokeModelWithResponseStream, and bedrock:ListFoundationModels
  • Amazon S3 — s3:GetObject, s3:PutObject, and s3:ListBucket on the relevant buckets
Follow the principle of least privilege — grant only the permissions your Digital Twin needs.

​
API Keys

By default, your Digital Twin uses Praxis-managed credits to access AI models. When you provide your own API keys, the Digital Twin routes requests directly to that provider using your account and token pool — giving you full control over billing, rate limits, and model availability. This is especially useful for organizations that already have enterprise agreements with AI providers or want to manage usage and costs independently.
Connects to OpenAI models including GPT-5.2, GPT-5.1, GPT-5-mini, o4-mini, GPT-Image-1.5, and Whisper transcription. Also enables OpenAI-powered features such as image generation and GPT-Realtime voice (Convo Mode).Generate your key at platform.openai.com/api-keys.
Connects to Anthropic Claude models including Claude Opus, Sonnet, and Haiku. Enables extended thinking and reasoning capabilities when available.Generate your key at console.anthropic.com/settings/keys.
Connects to Google Gemini models including Gemini 3.1 Pro, Gemini 3 Flash, Gemini 2.5 Flash, and Gemini Live for real-time voice. Uses the Google GenAI SDK directly for native tool calling and multimodal support.Generate your key at aistudio.google.com/apikey.
Connects to xAI Grok models including Grok-4, Grok-4 Fast Reasoning, and Grok Code Fast. Provides access to Grok’s reasoning and real-time knowledge capabilities.Generate your key at the xAI Console.
Connects to Mistral AI models via the native Mistral SDK — including Mistral Large, Medium, Small, Pixtral (vision), Magistral (reasoning), Codestral (code), Voxtral (TTS/STT), and embedding/moderation models.Generate your key at console.mistral.ai/api-keys.
Connects to DeepSeek models including DeepSeek-V3 and DeepSeek-R1 (reasoning). Provides high-performance models with competitive pricing.Generate your key at platform.deepseek.com/api_keys.
To enable this field for your institution, contact the Praxis AI team at humans@praxis-ai.com.
Connects to Cohere models including Command A (command-a-03-2025). Enables Cohere’s retrieval-augmented generation and enterprise search capabilities.Generate your key at dashboard.cohere.com/api-keys.
The same section also accepts keys for Stability AI (image/audio generation), OpenRouter, and InferX, plus the voice/avatar providers ElevenLabs, Anam, and LemonSlice — see the ElevenLabs section above and Realtime Voice & Avatars for avatar configuration.
How it works: When an API key is configured for a provider, any AI model from that provider selected in your AI Models configuration will use your key instead of Praxis-managed credits. If no key is set, the system falls back to the platform’s shared pool.
Security: All API keys are stored encrypted and excluded from API responses by default. They are never exposed in client-side code or logs. Only administrators with the appropriate entitlements can view or modify these values.