LLM Agents PHP SDK

LLM Agents is a PHP library for building and managing Language Model (LLM) based agents. It provides a framework for creating autonomous agents that can perform complex tasks, make decisions, and interact with various tools and APIs.

The library enables developers to integrate LLM capabilities into PHP applications efficiently, allowing for the creation of intelligent systems that can understand and respond to user inputs, process information, and carry out actions based on that processing.The library enables developers to integrate LLM capabilities into PHP applications efficiently.

For a comprehensive explanation of LLM agents and their applications, you can read the article: A PHP dev's dream: A PHP dev’s dream: An AI home that really gets you

For a complete example with sample agents and a CLI interface to interact with them, check out our sample application repository https://github.com/llm-agents-php/sample-app.

This sample app demonstrates practical implementations and usage patterns of the LLM Agents library.

The package does not include any specific LLM implementation. Instead, it provides a framework for creating agents that can interact with any LLM service or API.

✨ Key Features

🤖 Agent Creation: Create and configure LLM-based agents in PHP with customizable behaviors.
🔧 Tool Integration: Seamlessly integrate various tools and APIs for agent use in PHP applications.
🧠 Memory Management: Support for agent memory, enabling information retention and recall across interactions.
💡 Prompt Management: Efficient handling of prompts and instructions to guide agent behavior.
🔌 Extensible Architecture: Easily add new agent types, tools, and capabilities to your PHP projects.
🤝 Multi-Agent Support: Build systems with multiple interacting agents for complex problem-solving scenarios in PHP.

📀 Installation

You can install the LLM Agents package via Composer:

composer require llm-agents/agents

💻 Usage

→ Creating an Agent

To create an agent, you'll need to define its behavior, tools, and configuration. Here's a basic example:

use LLM\Agents\Agent\AgentAggregate;
use LLM\Agents\Agent\Agent;
use LLM\Agents\Solution\Model;
use LLM\Agents\Solution\ToolLink;
use LLM\Agents\Solution\MetadataType;
use LLM\Agents\Solution\SolutionMetadata;

class SiteStatusCheckerAgent extends AgentAggregate
{
    public const NAME = 'site_status_checker';

    public static function create(): self
    {
        $agent = new Agent(
            key: self::NAME,
            name: 'Site Status Checker',
            description: 'This agent checks the online status of websites.',
            instruction: 'You are a website status checking assistant. Your goal is to help users determine if a website is online. Use the provided tool to check site availability. Give clear, concise responses about a site\'s status.',
        );

        $aggregate = new self($agent);

        $aggregate->addMetadata(
            new SolutionMetadata(
                type: MetadataType::Memory,
                key: 'check_availability',
                content: 'Always check the site\'s availability using the provided tool.',
            ),
            new SolutionMetadata(
                type: MetadataType::Configuration,
                key: 'max_tokens',
                content: 500,
            )
        );

        $model = new Model(model: 'gpt-4o-mini');
        $aggregate->addAssociation($model);

        $aggregate->addAssociation(new ToolLink(name: CheckSiteAvailabilityTool::NAME));

        return $aggregate;
    }
}

→ Implementing a Tool

Now, let's implement the tool used by this agent:

use LLM\Agents\Tool\PhpTool;
use LLM\Agents\Tool\ToolLanguage;

class CheckSiteAvailabilityTool extends PhpTool
{
    public const NAME = 'check_site_availability';

    public function __construct()
    {
        parent::__construct(
            name: self::NAME,
            inputSchema: CheckSiteAvailabilityInput::class,
            description: 'This tool checks if a given URL is accessible and returns its HTTP status code and response time.',
        );
    }

    public function getLanguage(): ToolLanguage
    {
        return ToolLanguage::PHP;
    }

    public function execute(object $input): string
    {
        $ch = curl_init($input->url);
        curl_setopt_array($ch, [
            CURLOPT_RETURNTRANSFER => true,
            CURLOPT_HEADER => true,
            CURLOPT_NOBODY => true,
            CURLOPT_FOLLOWLOCATION => true,
            CURLOPT_MAXREDIRS => 10,
            CURLOPT_TIMEOUT => 30,
        ]);

        $startTime = microtime(true);
        $response = curl_exec($ch);
        $endTime = microtime(true);

        $statusCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
        $responseTime = round(($endTime - $startTime) * 1000, 2);

        curl_close($ch);

        $isOnline = $statusCode >= 200 && $statusCode < 400;

        return json_encode([
            'status_code' => $statusCode,
            'response_time_ms' => $responseTime,
            'is_online' => $isOnline,
        ]);
    }
}

And the input schema for the tool:

use Spiral\JsonSchemaGenerator\Attribute\Field;

class CheckSiteAvailabilityInput
{
    public function __construct(
        #[Field(title: 'URL', description: 'The full URL of the website to check')]
        public readonly string $url,
    ) {}
}

→ Linking Agents

LLM Agents supports creating complex systems by linking multiple agents together. This allows you to build hierarchical or collaborative agent networks. Here's how you can link one agent to another:

Creating an Agent Link

To link one agent to another, you use the AgentLink class. Here's an example of how to modify our SiteStatusCheckerAgent to include a link to another agent:

use LLM\Agents\Solution\AgentLink;

class SiteStatusCheckerAgent extends AgentAggregate
{
    public const NAME = 'site_status_checker';

    public static function create(): self
    {
        // ... [previous agent setup code] ...

        // Link to another agent
        $aggregate->addAssociation(
            new AgentLink(
                name: 'network_diagnostics_agent',
                outputSchema: NetworkDiagnosticsOutput::class,
            ),
        );

        return $aggregate;
    }
}

In this example, we're linking a network_diagnostics_agent. The outputSchema parameter specifies the expected output format from the linked agent. The output schema is used to standardize the data format that should be returned by the linked agent.

Using a Linked Agent

We don't provide an implementation for the linked agent here, but you can use the linked agent in your agent's execution.

Here's an example of how you might call the linked agent:

use LLM\Agents\Tool\PhpTool;
use LLM\Agents\Agent\AgentExecutor;
use LLM\Agents\LLM\Prompt\Chat\ToolCallResultMessage;
use LLM\Agents\LLM\Response\ToolCalledResponse;
use LLM\Agents\Tool\ToolExecutor;
use LLM\Agents\Tool\ToolLanguage;

/**
 * @extends PhpTool<AskAgentInput>
 */
final class AskAgentTool extends PhpTool
{
    public const NAME = 'ask_agent';

    public function __construct(
        private readonly AgentExecutor $executor,
        private readonly ToolExecutor $toolExecutor,
    ) {
        parent::__construct(
            name: self::NAME,
            inputSchema: AskAgentInput::class,
            description: 'Ask an agent with given name to execute a task.',
        );
    }

    public function getLanguage(): ToolLanguage
    {
        return ToolLanguage::PHP;
    }

    public function execute(object $input): string|\Stringable
    {
        $prompt = \sprintf(
            <<<'PROMPT'
%s
Important rules:
- Think before responding to the user.
- Don not markup the content. Only JSON is allowed.
- Don't write anything except the answer using JSON schema.
- Answer in JSON using this schema:
%s
PROMPT
            ,
            $input->question,
            $input->outputSchema,
        );

        while (true) {
            $execution = $this->executor->execute($input->name, $prompt);
            $result = $execution->result;
            $prompt = $execution->prompt;

            if ($result instanceof ToolCalledResponse) {
                foreach ($result->tools as $tool) {
                    $functionResult = $this->toolExecutor->execute($tool->name, $tool->arguments);

                    $prompt = $prompt->withAddedMessage(
                        new ToolCallResultMessage(
                            id: $tool->id,
                            content: [$functionResult],
                        ),
                    );
                }

                continue;
            }

            break;
        }

        return \json_encode($result->content);
    }
}