Show HN：Claude PHP 的開源 AI 代理框架

Hacker News·4 個月前

Hacker News 上展示了一個名為 Claude PHP Agents 的開源框架，專為使用 Claude 建構 AI 代理而設計。該框架支援 ReAct 循環、工具協調和層級式代理。

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

To see all available qualifiers, see our documentation.

Claude PHP Agents Framework

License

Uh oh!

There was an error while loading. Please reload this page.

claude-php/claude-php-agent

Folders and files

Latest commit

History

Repository files navigation

Claude PHP Agent Framework

A powerful PHP framework for building AI agents with Claude, featuring ReAct loops, tool orchestration, hierarchical agents, and advanced agentic patterns.

Features

Installation

Quick Start

Progress Updates (Streaming UI / Live Status)

While an agent is running, you can receive continuous progress events (iteration output, tool results, and streaming deltas when using StreamingLoop) via onUpdate():

Core Concepts

Loop Strategies

The framework provides multiple loop strategies for different types of tasks:

The Reason-Act-Observe pattern for general-purpose tasks:

Plan first, then execute systematically for complex multi-step tasks:

Generate, reflect, and refine for high-quality outputs:

See the Loop Strategies Guide for detailed documentation.

Tools

Tools give Claude the ability to interact with the world:

Agent Patterns

Memory & State

Production Features

⚡ NEW: Solve million-step tasks with near-zero error rates!

Key Features:

Paper Results: Successfully solved 20-disk Towers of Hanoi (1,048,575 moves) with ZERO errors!

See MakerAgent Documentation for detailed documentation.

🎯 NEW: Intelligent agent selection with automatic validation and adaptation!

Key Features:

See docs/adaptive-agent-service.md for detailed documentation.

Agent Patterns Reference

Configuration

Async & Concurrent Execution

The framework leverages AMPHP for true asynchronous and concurrent execution:

Batch Processing

Process multiple agent tasks concurrently:

Parallel Tool Execution

Execute multiple tool calls simultaneously:

Promise-Based Workflows

Use promises for async operations:

See the examples directory for complete async/concurrent examples.

Output Parsers

Transform unstructured LLM responses into structured data:

Available Parsers:

See Parsers Documentation for complete guide.

Examples

See the examples directory for 70+ complete working examples including:

Documentation

🎓 Getting Started

New to AI agents? Start with our comprehensive tutorial series:

📖 Complete Documentation

Requirements

Installation

For detailed setup instructions, see QUICKSTART.md.

Contributing

We welcome contributions! Please see: