General Settings

The General Settings section defines the global behavior of the AI Content Generator, including active state, connection parameters, selected LLM providers, and advanced search features.

License Activation Dependency

This module relies on the central Webkul licensing framework. All configuration fields and buttons are hidden and disabled by default until the license key is successfully validated under the centralized Module License configuration page.

To access these settings, navigate to:

Navigation Path

Stores → Configuration → Webkul → AI Content Generator

Global Toggles & Config Options

The following options are available under General Settings:

FieldOptions / TypeDescription
Enable ModuleYes / NoGlobally enable or disable the AI generation buttons, configurations, and consumers.
AI Model TypePre Configured Model / LLM ProviderSelect whether to route generation requests to a custom local server or direct to cloud providers.
Enable Prompt TemplatesYes / NoWhen set to Yes, enables the mapping of custom prompt templates to descriptions and SEO properties.
CMS Page Prompt TemplateDropdownThe default prompt template used when generating CMS page body content.

Configuration Path A: LLM Provider

If AI Model Type is set to LLM Provider, requests are processed using the Symfony AI bridge libraries:

  1. LLM Provider: Choose your preferred cloud engine:
    • OpenAI (latest available models)
    • Gemini (latest available models)
    • Anthropic (latest available models)
    • OpenRouter (OpenRouter multi-model proxy)
    • Cerebras (Cerebras inference engines)
    • DeepSeek (DeepSeek Chat/Coder engines)
  2. LLM API Key: Enter your secret developer API key for authentication. This is saved as an encrypted value in the database.
  3. Content Generation Model: Select the specific model to use from your provider's available list.
  4. Max Tokens: An integer specifying the maximum length of generated content response from the LLM.
  5. Enable Web Search: A yes/no switch that enables real-time web search capability during generation.

Web Search Tool Integration Details

The Enable Web Search (web_search_tool) option leverages provider-specific tools to retrieve external context from the web to augment generated content.

[!NOTE] Web search retrieval is only supported for OpenAI and Gemini models. Ensure your API key has permission to call these tools.

  • Google Gemini: Uses the Gemini server tool google_search => true to perform grounding search results using Google Search.
  • OpenAI: Employs OpenAI's standard web_search tool parameters.

General Configuration Panel

Configuration Path B: Pre Configured Model

If AI Model Type is set to Pre Configured Model, requests are sent to a dedicated self-hosted AI server endpoint (e.g. FastAPI / Python backend hosting custom model pipelines):

  1. Server Endpoint URL: The API host endpoint URL (e.g., http://your-ai-host.com:8000).
  2. Server API Key: Authentication token required to query the endpoint.
  3. Preconfigured Model: After clicking Fetch Models to query your server, select the loaded model identifier from the dropdown.

General Configuration Panel