Configuration 

The extension uses a database-based configuration architecture with three levels: Providers, Models, and Configurations. All management is done through the TYPO3 backend module.

Backend module 

Access the LLM management module at Admin Tools > LLM.

The backend module provides four sections:

Dashboard
Overview of registered providers, models, and configurations with status indicators.
Providers
Manage API connections with encrypted credentials. Test connections directly from the interface.
Models
Define available models with their capabilities and pricing. Fetch models from provider APIs.
Configurations
Create use-case-specific configurations with prompts and parameters.

Provider configuration 

Providers represent API connections with credentials. Create providers in Admin Tools > LLM > Providers.

Required fields 

identifier

identifier
Type
string
Required

true

Unique slug for programmatic access (e.g., openai-prod, ollama-local).

name

name
Type
string
Required

true

Display name shown in the backend.

adapter_type

adapter_type
Type
string
Required

true

The protocol to use. Available options:

  • openai - OpenAI API.
  • anthropic - Anthropic Claude API.
  • gemini - Google Gemini API.
  • ollama - Local Ollama instance.
  • openrouter - OpenRouter multi-model API.
  • mistral - Mistral AI API.
  • groq - Groq inference API.
  • azure_openai - Azure OpenAI Service.
  • custom - Custom OpenAI-compatible endpoint.

api_key

api_key
Type
string
Required

true

API key for authentication. Encrypted at rest using sodium_crypto_secretbox. Not required for local providers like Ollama.

Optional fields 

endpoint_url

endpoint_url
Type
string
Default
(adapter default)

Custom API endpoint. Leave empty to use the adapter's default URL.

organization_id

organization_id
Type
string
Default
(empty)

Organization ID for providers that support it (OpenAI, Azure).

timeout

timeout
Type
integer
Default
30

Request timeout in seconds.

max_retries

max_retries
Type
integer
Default
3

Number of retry attempts on failure.

options

options
Type
JSON
Default
{}

JSON object with additional adapter-specific options.

Testing provider connections 

Use the Test Connection button to verify provider configuration. The test makes an actual HTTP request to the provider's API and returns:

  • Connection status (success/failure).
  • Available models (if supported by the provider).
  • Error details (on failure).

Model configuration 

Models represent specific LLM models available through a provider. Create models in Admin Tools > LLM > Models.

Required fields 

identifier (model)

identifier (model)
Type
string
Required

true

Unique slug (e.g., gpt-5, claude-sonnet).

name (model)

name (model)
Type
string
Required

true

Display name (e.g., GPT-5 (128K)).

provider

provider
Type
reference
Required

true

Reference to the parent provider.

model_id

model_id
Type
string
Required

true

The API model identifier. Examples vary by provider:

  • OpenAI: gpt-5, gpt-5.2-instant, o4-mini.
  • Anthropic: claude-opus-4-5-20251101, claude-sonnet-4-5-20251101.
  • Google: gemini-3-pro-preview, gemini-3-flash-preview.

Optional fields 

context_length

context_length
Type
integer
Default
(provider default)

Maximum context window in tokens (e.g., 128000 for GPT-5).

max_output_tokens

max_output_tokens
Type
integer
Default
(model default)

Maximum output tokens (e.g., 16384).

capabilities

capabilities
Type
string (CSV)
Default
chat

Comma-separated list of supported features:

  • chat - Chat completion.
  • completion - Text completion.
  • embeddings - Text-to-vector.
  • vision - Image analysis.
  • streaming - Real-time streaming.
  • tools - Function/tool calling.

cost_input

cost_input
Type
integer
Default
0

Cost per 1M input tokens in cents (for cost tracking).

cost_output

cost_output
Type
integer
Default
0

Cost per 1M output tokens in cents.

is_default

is_default
Type
boolean
Default
false

Mark as default model for this provider.

Fetching models from providers 

Use the Fetch Models action to automatically retrieve available models from the provider's API. This populates the model list with the provider's current offerings.

LLM configuration 

Configurations define specific use cases with model selection and parameters. Create configurations in Admin Tools > LLM > Configurations.

Required fields 

identifier (config)

identifier (config)
Type
string
Required

true

Unique slug for programmatic access (e.g., blog-summarizer).

name (config)

name (config)
Type
string
Required

true

Display name (e.g., Blog Post Summarizer).

model

model
Type
reference
Required

true

Reference to the model to use.

system_prompt

system_prompt
Type
text
Required

true

System message that sets the AI's behavior and context.

Optional fields 

temperature

temperature
Type
float
Default
0.7

Creativity level from 0.0 (deterministic) to 2.0 (creative).

max_tokens (config)

max_tokens (config)
Type
integer
Default
(model default)

Maximum response length in tokens.

top_p

top_p
Type
float
Default
1.0

Nucleus sampling parameter (0.0 - 1.0).

frequency_penalty

frequency_penalty
Type
float
Default
0.0

Reduces word repetition (-2.0 to 2.0).

presence_penalty

presence_penalty
Type
float
Default
0.0

Encourages topic diversity (-2.0 to 2.0).

use_case_type

use_case_type
Type
string
Default
chat

The type of task:

  • chat - Conversational interactions.
  • completion - Text completion.
  • embedding - Vector generation.
  • translation - Language translation.

Using configurations 

Retrieve configurations programmatically:

Example: Using configurations in a controller 
use Netresearch\NrLlm\Domain\Repository\LlmConfigurationRepository;
use Netresearch\NrLlm\Provider\ProviderAdapterRegistry;

class MyController
{
    public function __construct(
        private readonly LlmConfigurationRepository $configRepository,
        private readonly ProviderAdapterRegistry $adapterRegistry,
    ) {}

    public function processAction(): void
    {
        // Get configuration by identifier
        $config = $this->configRepository->findByIdentifier('blog-summarizer');

        // Get the model and provider
        $model = $config->getModel();
        $provider = $model->getProvider();

        // Create adapter and make requests
        $adapter = $this->adapterRegistry->createAdapterFromModel($model);
        $response = $adapter->chatCompletion($messages, $config->toOptions());
    }
}
Copied!

TypoScript settings 

Runtime settings can be configured via TypoScript:

Constants 

Configuration/TypoScript/constants.typoscript 
plugin.tx_nrllm {
    settings {
        # Default temperature (0.0-2.0)
        defaultTemperature = 0.7

        # Maximum tokens for responses
        defaultMaxTokens = 1000

        # Cache lifetime in seconds
        cacheLifetime = 3600

        # Enable/disable response caching
        enableCaching = 1

        # Enable streaming by default
        enableStreaming = 0
    }
}
Copied!

Environment variables 

For deployment flexibility, use environment variables:

.env 
# TYPO3 encryption key (used for API key encryption)
TYPO3_CONF_VARS__SYS__encryptionKey=your-secure-encryption-key

# Optional: Override provider settings via environment
TYPO3_NR_LLM_DEFAULT_TIMEOUT=60
Copied!

Security 

API key protection 

  1. Encrypted storage: API keys are encrypted using sodium_crypto_secretbox.
  2. Database security: Ensure database backups are encrypted.
  3. Backend access: Restrict backend module access to authorized users.
  4. Key rotation: Changing the TYPO3 encryptionKey requires re-encryption.

Input sanitization 

Always sanitize user input before sending to LLM providers:

Example: Sanitizing user input 
use TYPO3\CMS\Core\Utility\GeneralUtility;

$sanitizedInput = GeneralUtility::removeXSS($userInput);
$response = $adapter->chatCompletion([
    ['role' => 'user', 'content' => $sanitizedInput]
]);
Copied!

Output handling 

Treat LLM responses as untrusted content:

Example: Escaping output 
$response = $adapter->chatCompletion($messages);
$safeOutput = htmlspecialchars($response->content, ENT_QUOTES, 'UTF-8');
Copied!

Logging 

Enable detailed logging for debugging:

config/system/additional.php 
$GLOBALS['TYPO3_CONF_VARS']['LOG']['Netresearch']['NrLlm'] = [
    'writerConfiguration' => [
        \Psr\Log\LogLevel::DEBUG => [
            \TYPO3\CMS\Core\Log\Writer\FileWriter::class => [
                'logFileInfix' => 'nr_llm',
            ],
        ],
    ],
];
Copied!

Log file location: var/log/typo3_nr_llm_*.log

Caching 

The extension uses TYPO3's caching framework:

  • Cache identifier: nrllm_responses.
  • Default TTL: 3600 seconds (1 hour).
  • Embeddings TTL: 86400 seconds (24 hours).

Clear cache via CLI:

Clear extension caches 
vendor/bin/typo3 cache:flush --group=nrllm
Copied!