> For the complete documentation index, see [llms.txt](https://docs.infraon.io/infraon-help/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.infraon.io/infraon-help/infinity-admin-guide/infraon-platform/feature-configuration.md).

# Feature Configuration

The Feature Configuration module enables administrators to manage and control **system features and AI capabilities** across the Infraon Platform through a centralized interface.

It allows configuration of feature behavior without requiring backend changes, ensuring consistency across modules and services. Administrators can track, discover, and enable all AI-related capabilities and privileges available within the organization.

The module provides granular control over how AI-driven functionalities operate, including knowledge base generation, search behavior, automation intelligence, and user experience settings.

Available sections include:

* Knowledge Base Configuration: Configure AI-driven knowledge base creation, accuracy, and processing parameters.
* Ticket AI: Manage AI capabilities related to ticket analysis, suggestions, and automation.
* Request AI: Configure AI behavior for request classification and processing.
* Problem AI: Control AI-assisted problem identification and root cause insights.
* Change AI: Manage AI recommendations and validations for change management workflows.
* Release AI: Configure AI support for release planning and execution insights.
* AI Service Classification: Define how AI classifies services based on input data.
* Bot Assistants: Manage AI-powered virtual assistants and their behavior.
* SLA Prediction: Configure AI-based SLA prediction and risk analysis settings.
* Similar Ticket Merge: Control AI logic for identifying and merging similar tickets.
* Agentic Bot: Configure autonomous AI agents for task execution and assistance.

## **Data Sanitization Configuration**

Users can automatically detect and sanitize **sensitive information** before AI-powered features process the data. The configuration helps protect **confidential information** such as personal identifiers, credentials, financial details, network information, and authentication tokens by masking, replacing, or removing sensitive values before they are sent to the AI model.

### **How does it work**

The sanitization engine scans all incoming data as plain text and applies **regex-based pattern** detection to identify sensitive information across multiple supported categories. Based on the configured settings, detected values are either replaced with placeholders, substituted with synthetic values, or removed completely before being sent to the AI model.

When **Send Faker Data** to AI Model is enabled, the system replaces detected sensitive values with realistic synthetic substitutes while maintaining a reversible mapping. This allows the AI model to retain contextual understanding without exposing actual sensitive information.

![](/files/1a2a686a59989df003244ce25412b5071e06e1bd)

Once the AI response is generated, the original values are automatically restored before the response is displayed to the user.

**Configurations**

**General Settings |** Basic Details

<table><thead><tr><th width="152.199951171875">Label</th><th width="357">Description</th><th>Example</th></tr></thead><tbody><tr><td><strong>Enable Data Sanitization</strong></td><td>Enables or disables sanitization for all AI requests before the AI model processes the data.</td><td>When enabled, sensitive information is sanitized before being sent to AI services.</td></tr><tr><td><strong>Send Faker Data to AI Model</strong></td><td>Replaces sensitive values with realistic synthetic substitutes instead of placeholders. Original values are automatically restored in AI responses via reversible mapping.</td><td>john@company.com → user1@sanitizer.example</td></tr></tbody></table>

**Sanitization Categories |** Basic Details

Sanitization Categories allow users to control which types of sensitive information should be detected and sanitized before data is processed by AI features. Each category can be enabled or disabled individually based on organizational privacy and compliance requirements.

![](/files/478caee161bc57d05f490b6abea4a5a22763a5a5)

<table><thead><tr><th width="186.5999755859375">Label</th><th>Description</th><th>Example</th></tr></thead><tbody><tr><td><strong>Enable All Categories</strong></td><td>Enables or disables all sanitization category toggles together.</td><td>Enabling this option automatically enables all category-based sanitization controls.</td></tr><tr><td><strong>MAC Address</strong></td><td>Redacts MAC addresses in colon, dash, and Cisco dot formats.</td><td>AA:BB:CC:DD:EE:FF → [MAC_ADDRESS]</td></tr><tr><td><strong>IP Address</strong></td><td>Redacts IPv4, IPv4 with CIDR notation, and IPv6 addresses.</td><td>192.168.1.1 → [IP_ADDRESS]</td></tr><tr><td><strong>Email Address</strong></td><td>Redacts standard and obfuscated email addresses.</td><td>user@domain.com → [EMAIL]</td></tr><tr><td><strong>Phone Numbers</strong></td><td>Redacts international and local phone number formats, including E.164, Indian, UK, and North American formats.</td><td>+91 98765 43210 → [PHONE]</td></tr><tr><td><strong>Credit / Debit Card</strong></td><td>Redacts card numbers, CVV/CVC values, and expiry dates for supported card providers.</td><td>4111 1111 1111 1111 → [CARD_NUMBER]</td></tr><tr><td><strong>National / Gov. IDs</strong></td><td>Redacts government-issued identity numbers such as Aadhaar, PAN, Passport, SSN, GSTIN, and Driving License numbers.</td><td>ABCDE1234F → [PAN]</td></tr><tr><td><strong>Date of Birth &#x26; Age</strong></td><td>Redacts date of birth information and age references in supported formats.</td><td>DOB: 15/08/1990 → [DOB]</td></tr><tr><td><strong>Secrets &#x26; API Keys</strong></td><td>Redacts API keys, tokens, passwords, JWTs, SSH keys, database connection strings, and other secret values.</td><td>sk-abc123XYZ → [API_KEY]</td></tr><tr><td><strong>URL Credentials</strong></td><td>Redacts credentials and sensitive query parameters embedded within URLs.</td><td>https://user:pass@host.com → https://[CREDENTIALS]@[SERVER]</td></tr><tr><td><strong>Banking Identifiers</strong></td><td>Redacts IFSC, IBAN, SWIFT/BIC, UPI IDs, routing numbers, and bank account numbers.</td><td>HDFC0001234 → [IFSC]</td></tr><tr><td><strong>Server &#x26; Hostnames</strong></td><td>Redacts fully qualified domain names (FQDNs), Windows machine names, and server hostnames.</td><td>prod-server-01.internal.local → [SERVER]</td></tr><tr><td><strong>Location &#x26; Coordinates</strong></td><td>Redacts GPS coordinates, DMS coordinates, Google Plus Codes, and address details.</td><td>12.9716, 77.5946 → [GEO_COORDINATES]</td></tr><tr><td><strong>Vehicle Registration</strong></td><td>Redacts vehicle registration numbers and VIN details.</td><td>MH12AB1234 → [VEHICLE_REG]</td></tr><tr><td><strong>Session &#x26; Auth Tokens</strong></td><td>Redacts session identifiers, authentication headers, cookies, CSRF tokens, and OAuth parameters.</td><td>JSESSIONID=ABC123XYZ → [SESSION_ID]</td></tr></tbody></table>

## **AI Knowledge Base**

The Knowledge Base Configuration section allows administrators to control how **AI-generated** knowledge base articles are created, processed, and presented across the platform. It provides:

* Granular settings to manage content quality
* Search behavior
* AI response handling
* System performance
* Data privacy

Administrators can enable or disable the Knowledge Base feature, fine-tune AI thresholds, control KB suggestions in search and email responses, and define limits on content structure. These settings directly impact how knowledge is generated, matched, and delivered to users.

To know more, refer to the [Knowledge Base](/infraon-help/infinity-user-guide/knowledge-base.md) module.

### **Configuration**

#### **Enable Knowledge Base**

<table><thead><tr><th width="160.20001220703125">Label</th><th width="81.79998779296875">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Enable Knowledge Base</strong></td><td>Toggle</td><td>Enables or disables the Knowledge Base feature across the platform.<br><br><strong>When enabled,</strong> the system enables AI-driven KB creation, similarity-based retrieval, and the attachment of KB articles to tickets and emails.<br><br><strong>When disabled,</strong> all KB-related AI operations, such as generation, search, and attachment, are restricted.</td></tr></tbody></table>

#### **KB Quality Settings**

<table><thead><tr><th width="169">Label</th><th width="104.99993896484375">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>KB Creation Standard</strong></td><td>Input</td><td>Defines the minimum quality or eligibility score required for a ticket to be converted into a KB article. This ensures that only meaningful and well-resolved tickets are considered for KB creation. Lower values increase the volume of KB creation, while higher values enforce stricter quality control.</td></tr><tr><td><strong>Similarity Threshold</strong></td><td>Input</td><td>Determines how closely a new ticket or query must match an existing KB article for it to be reused or updated. Higher thresholds ensure more accurate matches but may reduce reuse, while lower thresholds allow broader matching.</td></tr><tr><td><strong>AI Context Matching</strong></td><td>Input</td><td>Controls how strictly the AI evaluates contextual similarity (problem description, resolution steps, metadata) during KB generation and retrieval. Higher values improve contextual accuracy but may limit matches.</td></tr></tbody></table>

#### **User Experience Settings**

<table><thead><tr><th width="151.39996337890625">Label</th><th width="94.800048828125">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Enable Smart KB Suggestions</strong></td><td>Toggle</td><td>Enables AI-powered semantic search to suggest relevant KB articles during ticket creation, updates, and search operations. When enabled, users receive contextual recommendations based on issue descriptions and activity.</td></tr><tr><td><strong>Search Relevance</strong></td><td>Input</td><td>Defines the relevance score threshold for displaying KB articles in search results. Higher values ensure only highly relevant KBs are shown, while lower values allow more results with broader matching.</td></tr><tr><td><strong>Search Results Count</strong></td><td>Input</td><td>Specifies the maximum number of KB articles returned in search results. This helps control UI clutter and ensures users see a manageable list of relevant KBs.</td></tr><tr><td><strong>Email Attachments Limit</strong></td><td>Input</td><td>Sets the maximum number of KB articles that can be attached to ticket response emails. The system prioritizes the most relevant KBs based on similarity score and attaches them up to the defined limit.</td></tr></tbody></table>

<figure><img src="/files/TKaVw8gcKFT8o55B0mRg" alt=""><figcaption></figcaption></figure>

#### **AI Behavior Settings**

<table><thead><tr><th width="133">Label</th><th width="93.39996337890625">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Response Consistency</strong></td><td>Input</td><td>Controls the variability of AI-generated content.<br><br><strong>Lower values</strong> produce more deterministic, consistent responses, while higher values allow greater variation and creativity in generated KB content and responses.</td></tr></tbody></table>

#### **AI Technical Settings**

<table><thead><tr><th width="151.4000244140625">Label</th><th width="78">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Max Retries</strong></td><td>Input</td><td>Defines the number of retry attempts for failed AI or API requests. This ensures temporary failures (e.g., network issues or provider delays) are retried before marking the operation as failed.</td></tr><tr><td><strong>Max Tokens Per Request</strong></td><td>Input</td><td>Limits the number of tokens processed per AI request. This helps control response size, processing cost, and performance, especially when handling large KB content.</td></tr><tr><td><strong>Processing Timeout</strong></td><td>Input</td><td>Specifies the maximum time (in seconds) the system waits for an AI response before timing out. If exceeded, the operation is stopped and handled as a failure or fallback scenario.</td></tr></tbody></table>

#### **Data Privacy Settings**

<table><thead><tr><th width="150.5999755859375">Label</th><th width="101.99993896484375">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Remove Personal Information</strong></td><td>Toggle</td><td><strong>Enables</strong> removal of personally identifiable information (PII) from KB content. <br><br>When <strong>disabled,</strong> all related privacy filters are disabled as well, allowing the original data to remain in KB articles.</td></tr><tr><td><strong>Remove Email Addresses</strong></td><td>Toggle</td><td>Automatically detects and removes email addresses from KB content to prevent exposure of sensitive user or organizational data.</td></tr><tr><td><strong>Remove Phone Numbers</strong></td><td>Toggle</td><td>Removes phone numbers from KB articles to ensure privacy and compliance with data protection policies.</td></tr><tr><td><strong>Remove IP Addresses</strong></td><td>Toggle</td><td>Filters out IP addresses from KB content to prevent disclosure of infrastructure-level details.</td></tr><tr><td><strong>Remove Server Names</strong></td><td>Toggle</td><td>Removes server names and host identifiers from KB content to avoid exposing internal system details.</td></tr><tr><td><strong>Use Placeholders</strong></td><td>Toggle</td><td>Replaces sensitive information with placeholders such as <strong>[USER], [EMAIL], or [SERVER]</strong> rather than removing it completely, preserving context while maintaining privacy.</td></tr></tbody></table>

#### **KB Content Length Limits**

<table><thead><tr><th width="153">Label</th><th width="96.20001220703125">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Max Title Length</strong></td><td>Input</td><td>Defines the maximum allowed length for KB titles. This ensures titles remain concise, readable, and suitable for search indexing.</td></tr><tr><td><strong>Max Problem Length</strong></td><td>Input</td><td>Limits the length of the problem description section in KB articles to maintain clarity and avoid excessively long content.</td></tr><tr><td><strong>Max Solution Length</strong></td><td>Input</td><td>Defines the maximum length of the solution section to ensure structured, readable, and actionable resolution steps.</td></tr><tr><td><strong>Max Keywords Count</strong></td><td>Input</td><td>Specifies the maximum number of keywords that can be associated with a KB article, helping improve searchability without overloading metadata.</td></tr></tbody></table>

#### **Actions**

<table><thead><tr><th width="147.39996337890625">Label</th><th width="84">Action</th><th>Description/ Example</th></tr></thead><tbody><tr><td><strong>Save Settings</strong></td><td>Button</td><td>Saves all configured values and applies them across KB generation, search, and processing workflows. Changes take effect immediately after saving.</td></tr><tr><td><strong>Reset to Default</strong></td><td>Button</td><td>Resets all configuration fields to predefined system default values. Useful for reverting incorrect configurations or restoring recommended settings.</td></tr></tbody></table>

<figure><img src="/files/LvigyOJrzUbcDPWpY4EU" alt=""><figcaption></figcaption></figure>

## **Ticket – AI Configuration**

The Ticket AI Configuration section allows administrators to control how AI capabilities assist in ticket handling, response drafting, and communication improvements.

It enables automation of summaries and enhancement of ticket content to improve clarity and response efficiency.

For more information, refer to the [Ticket ](/infraon-help/infinity-user-guide/ticket-management.md)module.

### **Configuration**

<table><thead><tr><th width="173.79998779296875">Label</th><th width="89.39996337890625">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Conversation Summary [AI Generated]</strong></td><td>Toggle</td><td>Automatically generates a concise summary of ticket conversations, including interactions, updates, and resolutions.<br><br><strong>When enabled,</strong> users can quickly understand ticket context without reviewing full conversation threads.<br><br><strong>When disabled,</strong> summaries are not generated, and users must review the complete ticket history manually.</td></tr><tr><td><strong>Text Enhancement</strong></td><td>Toggle</td><td>Enhances <strong>ticket content with AI,</strong> improving grammar, clarity, tone, and structure.<br><br><strong>When enabled,</strong> users can rephrase, expand, or refine responses before sending.<br><br><strong>When disabled,</strong> responses remain unchanged and fully manual.</td></tr></tbody></table>

<figure><img src="/files/z5uKEALvvjvAe54VM5s7" alt=""><figcaption></figcaption></figure>

## **Request – AI Configuration**

The Request AI Configuration section enables AI-driven enhancements for service requests, improving communication clarity and providing summarized insights.

For more information, refer to the [Request ](/infraon-help/infinity-user-guide/request-management.md)module.

#### **Configuration**

<table><thead><tr><th width="163.4000244140625">Label</th><th width="79.199951171875">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Conversation Summary [AI Generated]</strong></td><td>Toggle</td><td>Generates a summary of request-related interactions, approvals, and updates.<br><br>Helps users quickly understand the progress of requests and decisions without having to review all activity logs.</td></tr><tr><td><strong>Text Enhancement</strong></td><td>Toggle</td><td>Improves request descriptions and responses by rephrasing and structuring content for better readability and professionalism.<br><br>Ensures consistent and clear communication across request workflows.</td></tr></tbody></table>

<figure><img src="/files/coMHwEDjG6NSPSberlD2" alt=""><figcaption></figcaption></figure>

## **Problem – AI Configuration**

The Problem AI Configuration section controls AI assistance for problem records, enabling better analysis, documentation, and communication.

For more information, refer to the [Problem ](/infraon-help/infinity-user-guide/problem-management.md)module.

#### **Configuration**

<table><thead><tr><th width="153.79998779296875">Label</th><th width="93.79998779296875">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Conversation Summary [AI Generated]</strong></td><td>Toggle</td><td>Generates a summarized view of problem investigation, discussions, and resolutions.<br><br>Helps users quickly understand the overall problem lifecycle and key decisions.</td></tr><tr><td><strong>Text Enhancement</strong></td><td>Toggle</td><td>Enhances problem descriptions, root-cause details, and resolution notes through AI-based improvements.<br><br>Ensures clarity and consistency in problem documentation.</td></tr></tbody></table>

<figure><img src="/files/viCPMyJLi03zPzjhgW7Y" alt=""><figcaption></figcaption></figure>

## **Change – AI Configuration**

The Change AI Configuration section enables AI support for change records, improving documentation quality and clarity of communication.

To know more, refer to the [Change ](/infraon-help/infinity-user-guide/change-management.md)module.

#### **Configuration**

<table><thead><tr><th width="157.79998779296875">Label</th><th width="98.20001220703125">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Conversation Summary [AI Generated]</strong></td><td>Toggle</td><td>Summarizes change discussions, approvals, and implementation notes.<br><br>Provides a quick overview of the change lifecycle and decision points.</td></tr><tr><td><strong>Text Enhancement</strong></td><td>Toggle</td><td>Enhances change request details, implementation plans, and communication using AI-powered text improvements.<br><br>Helps maintain structured and clear change documentation.</td></tr></tbody></table>

<figure><img src="/files/jveQuVFvvjHkLfwvNAen" alt=""><figcaption></figcaption></figure>

## **Release – AI Configuration**

The Release AI Configuration section allows AI assistance in release documentation and communication.

For more information, refer to the [Release ](/infraon-help/infinity-user-guide/release-management.md)module.

### **Configuration**

<table><thead><tr><th width="139.4000244140625">Label</th><th width="105.5999755859375">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Conversation Summary [AI Generated]</strong></td><td>Toggle</td><td>Generates summaries of release activities, updates, and deployment discussions.<br><br>Helps users quickly review release progress and outcomes.</td></tr><tr><td><strong>Text Enhancement</strong></td><td>Toggle</td><td>Improves release notes, descriptions, and communication content using AI enhancements.<br><br>Ensures professional and clear release communication.</td></tr></tbody></table>

<figure><img src="/files/BzcR9P6XoCEzCybCTlim" alt=""><figcaption></figcaption></figure>

## **AI Service Classification**

The AI Service Classification section enables automated ticket and request classification using AI models.

To know more, refer to the [Service Classification](/infraon-help/infinity-user-guide/infraon-configuration/service-management.md) module.

### **Configuration**

<table><thead><tr><th width="141.800048828125">Label</th><th width="104.5999755859375">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Allow Process Classification</strong></td><td>Toggle</td><td>Automatically classifies incoming emails into tickets or requests using AI.<br><br>Reduces manual effort in identifying and routing incoming requests.</td></tr><tr><td><strong>Enable Service Classification</strong></td><td>Toggle</td><td>Enables AI-based service classification based on ticket/request content.<br><br>When enabled, AI dynamically assigns categories.<br><br>When disabled, classification must be done manually.</td></tr></tbody></table>

<figure><img src="/files/QMAuDIUgTaKOYTA2Ig9t" alt=""><figcaption></figcaption></figure>

## **Bot Assistants**

The Bot Assistants section enables AI-powered bots to provide automated responses and guidance to users.

For more information, refer to the [Bot Assistants](/infraon-help/infinity-user-guide/infraon-configuration/bots/bots-assistance.md) module.

### **Configuration**

<table><thead><tr><th width="196.20001220703125">Label</th><th width="81.39996337890625">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Enable Bot Assistants</strong></td><td>Toggle</td><td>Allows AI bots to provide automated responses, suggestions, and assistance across the platform.<br><br>Helps resolve common queries more quickly and reduces the manual workload for support teams.</td></tr></tbody></table>

<figure><img src="/files/BF2MhmONcoN5ZZVbq4Go" alt=""><figcaption></figcaption></figure>

## **SLA Prediction**

The SLA Prediction section enables AI to predict potential SLA breaches and delays.

For more information, refer to the [SLA](/infraon-help/infinity-user-guide/sla-management.md) module.

### **Configuration**

<table><thead><tr><th width="173.79998779296875">Label</th><th width="88.5999755859375">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Enable SLA Prediction</strong></td><td>Toggle</td><td>Uses AI to analyze ticket patterns and predict potential SLA breaches or delays.<br><br>Enables proactive action to maintain SLA compliance and avoid violations.</td></tr></tbody></table>

<figure><img src="/files/VknVIZgF7OHswKkNKfo3" alt=""><figcaption></figcaption></figure>

## **Similar Ticket Merge**

The Similar Ticket Merge section helps identify and manage duplicate or similar tickets using AI.

For more information, refer to the [Ticket ](/infraon-help/infinity-user-guide/ticket-management.md)module.

### **Configuration**

<table><thead><tr><th width="157.800048828125">Label</th><th width="94.20001220703125">Action</th><th>Description / Example</th></tr></thead><tbody><tr><td><strong>Enable AI Similar Ticket Merge</strong></td><td>Toggle</td><td>Automatically detects and merges duplicate or similar tickets based on content similarity.<br><br>Improves efficiency by reducing duplicate efforts and consolidating related issues.</td></tr><tr><td><strong>Rule-Based Similar Incident Sidebar</strong></td><td>Toggle</td><td>Displays similar incidents in a sidebar based on predefined rules when viewing a ticket.<br><br>Helps agents quickly identify related issues and reuse existing solutions.</td></tr></tbody></table>

<figure><img src="/files/2oITNotLywCD1SWfqIT9" alt=""><figcaption></figcaption></figure>