# Aigie
[](https://badge.fury.io/py/aigie)
[](https://www.python.org/downloads/)
[](https://opensource.org/licenses/MIT)
[](https://github.com/psf/black)
[](https://pytest.org/)
> **AI Agent Runtime Error Detection & Remediation**
Aigie is a real-time error detection and monitoring system for LangChain and LangGraph applications with **intelligent error remediation capabilities** and a revolutionary **LLM-as-Judge validation system**. It provides seamless integration without requiring additional code from users, automatically detecting, analyzing, validating, and fixing runtime errors as they occur.
## ✨ Features
- **🚀 Zero-Code Integration** - Automatically detects and wraps LangChain/LangGraph applications
- **⚡ Real-time Error Detection** - Immediate error reporting with classification and severity assessment
- **🧠 LLM-as-Judge Validation** - Revolutionary AI-powered step validation using 6 validation strategies
- **🤖 Gemini-Powered Analysis** - AI-powered error classification and intelligent remediation
- **🔄 Intelligent Retry System** - Automatic retry with enhanced context from Gemini
- **💉 Prompt Injection Remediation** - Actually fixes errors by injecting guidance into AI agent prompts
- **🔧 Auto-Correction System** - Automatically fixes invalid steps using multiple correction strategies
- **📊 Comprehensive Monitoring** - Covers execution, API, state, and memory errors
- **📈 Performance Insights** - Track execution time, memory usage, and resource consumption
- **🧠 Pattern Learning** - Learns from successful and failed operations to improve over time
## 🚀 Quick Start
### Installation
```bash
pip install aigie
```
### Basic Usage (Zero Code Changes)
```python
# Just import Aigie - it automatically starts monitoring
from aigie import auto_integrate
# Your existing LangChain code works unchanged
from langchain_core.prompts import PromptTemplate
from langchain_core.runnables import RunnableSequence
# Aigie automatically intercepts and monitors
prompt = PromptTemplate.from_template("Tell me a joke about {topic}")
chain = prompt | llm
# Run normally - Aigie monitors in background
result = chain.invoke({"topic": "programming"})
```
### LangGraph Integration
```python
# Your existing LangGraph code works unchanged
from langgraph.graph import StateGraph, END
# Aigie automatically monitors state transitions and node execution
graph = StateGraph(StateType)
# ... your graph setup ...
app = graph.compile()
# Run normally - Aigie monitors in background
result = app.invoke({"input": "Hello"})
```
## 🧠 LLM-as-Judge Validation System
Aigie's revolutionary **LLM-as-Judge** validation system provides real-time validation and correction of AI agent execution steps. This system ensures agents execute correctly and efficiently by continuously monitoring, validating, and automatically correcting their behavior.
### 🎯 Validation Strategies
The system uses **6 comprehensive validation strategies** to judge each execution step:
| Strategy | Description |
|----------|-------------|
| **Goal Alignment** | Does this step advance the agent's stated goal? |
| **Logical Consistency** | Is the step logically sound given the context? |
| **Output Quality** | Will this likely produce appropriate output? |
| **State Coherence** | Does this maintain consistent agent state? |
| **Safety Compliance** | Does this follow safety guidelines? |
| **Performance Optimality** | Is this the most efficient approach? |
### 🔧 Auto-Correction System
When validation fails, Aigie automatically attempts correction using multiple strategies:
- **Parameter Adjustment** - Fix incorrect parameters
- **Prompt Refinement** - Improve prompts and input data
- **Tool Substitution** - Replace wrong tools with correct ones
- **Logic Repair** - Fix logical errors in reasoning
- **Goal Realignment** - Align steps with agent goals
- **State Restoration** - Fix corrupted agent state
### 📊 Advanced Features
- **Parallel Validation** - Multiple strategies run simultaneously for faster processing
- **Pattern Learning** - Learns from validation history to improve future validations
- **Intelligent Caching** - Caches validation results for similar steps
- **Adaptive Thresholds** - Dynamically adjusts validation criteria based on performance
- **Rich Context Capture** - Captures comprehensive execution context for intelligent validation
### Example Usage
```python
from aigie import auto_integrate
from aigie.core.validation import ValidationEngine
# Auto-integration enables LLM-as-Judge validation automatically
auto_integrate()
# Your existing code works unchanged - validation happens automatically
from langchain_core.prompts import PromptTemplate
from langchain_core.runnables import RunnableSequence
prompt = PromptTemplate.from_template("Tell me a joke about {topic}")
chain = prompt | llm
# Each step is automatically validated and corrected if needed
result = chain.invoke({"topic": "programming"})
```
## 🤖 Gemini Integration
Aigie supports two ways to use Gemini:
### 1. Vertex AI (Recommended for production)
```bash
export GOOGLE_CLOUD_PROJECT=your-project-id
gcloud auth application-default login
gcloud services enable aiplatform.googleapis.com
```
### 2. Gemini API Key (Best for local/dev)
```bash
export GEMINI_API_KEY=your-gemini-api-key
# Get from https://aistudio.google.com/app/apikey
```
### Install Gemini dependencies
```bash
pip install google-cloud-aiplatform vertexai google-generativeai
```
## 🔧 Advanced Usage
### Intelligent Retry with Enhanced Context
```python
from aigie.core.intelligent_retry import intelligent_retry
@intelligent_retry(max_retries=3)
def my_function(input_data):
# If this fails, Aigie will:
# 1. Analyze the error with Gemini
# 2. Generate enhanced retry context
# 3. Automatically retry with better parameters
return process_data(input_data)
```
### CLI Usage
```bash
# Enable monitoring
aigie enable --config development
# Show status
aigie status
# Show detailed analysis
aigie analysis
# Gemini Integration
aigie gemini --setup your-project-id
aigie gemini --status
aigie gemini --test
```
## 📋 Error Types Detected
| Category | Description |
|----------|-------------|
| **Execution Errors** | Runtime exceptions, timeouts, infinite loops |
| **API Errors** | External service failures, rate limits, authentication issues |
| **State Errors** | Invalid state transitions, data corruption, type mismatches |
| **Memory Errors** | Overflow, corruption, persistence failures |
| **Performance Issues** | Slow execution, resource exhaustion, memory leaks |
| **Framework-specific** | LangChain chain/tool/agent errors, LangGraph node/state errors |
## 📊 Monitoring Capabilities
- **Real-time Error Logging** - Immediate error reporting with classification
- **Performance Metrics** - Execution time, memory usage, API call latency
- **State Tracking** - Monitor agent state changes and transitions
- **Resource Monitoring** - CPU, memory, and disk usage with health indicators
- **AI-Powered Analysis** - Intelligent error classification and remediation strategies
- **Pattern Learning** - Learns from successful and failed operations
## 🏗️ Project Structure
```
aigie/
├── core/ # Core functionality
│ ├── types/ # Type definitions and data structures
│ │ ├── error_types.py # Error classification and severity
│ │ └── validation_types.py # Validation data structures
│ ├── validation/ # 🧠 LLM-as-Judge validation system
│ │ ├── runtime_validator.py # LLM-as-Judge implementation
│ │ ├── step_corrector.py # Auto-correction system
│ │ ├── validation_engine.py # Main orchestrator
│ │ ├── validation_pipeline.py # Multi-stage validation
│ │ ├── validation_monitor.py # Performance monitoring
│ │ └── context_extractor.py # Context inference
│ ├── error_handling/ # Error detection and handling
│ │ ├── error_detector.py # Main error detection engine
│ │ └── intelligent_retry.py # Smart retry system
│ ├── monitoring/ # Performance and resource monitoring
│ │ └── monitoring.py # Resource monitoring
│ ├── ai/ # AI/LLM components
│ │ └── gemini_analyzer.py # Gemini-powered analysis
│ └── utils/ # Utility functions
├── interceptors/ # Framework-specific interceptors
│ ├── langchain.py # LangChain interceptor
│ ├── langgraph.py # LangGraph interceptor
│ └── validation_interceptor.py # Enhanced interceptor with validation
├── reporting/ # Error reporting and logging
├── utils/ # Utility functions
├── cli.py # Command-line interface
└── auto_integration.py # Automatic integration system
```
## ⚙️ Configuration
### Environment Variables
```bash
export AIGIE_LOG_LEVEL=INFO
export AIGIE_ENABLE_METRICS=true
export AIGIE_ERROR_THRESHOLD=5
export AIGIE_ENABLE_ALERTS=true
```
### Configuration Files
```bash
# Generate configuration
aigie config --generate config.yml
# Use configuration
aigie enable --config config.yml
```
## 🛠️ Development
### Prerequisites
- Python 3.9+
- pip
- git
### Setup Development Environment
```bash
# Clone the repository
git clone https://github.com/NirelNemirovsky/aigie-io.git
cd aigie-io
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
pip install -e .
```
### Running Tests
```bash
# Run all tests
pytest tests/ -v
# Run specific test categories
pytest tests/unit/ -v
pytest tests/integration/ -v
pytest tests/e2e/ -v
```
### Running Examples
```bash
# Set up Gemini (one-time)
export GOOGLE_CLOUD_PROJECT=your-project-id
# Run comprehensive example
python examples/ai_research_assistant.py
```
### Code Quality
```bash
# Format code
black aigie/ tests/ examples/
# Lint code
flake8 aigie/ tests/ examples/
# Type checking
mypy aigie/
```
## 📈 Current Status
✅ **Fully Implemented and Working**:
- **🧠 LLM-as-Judge Validation System** - Revolutionary AI-powered step validation with 6 validation strategies
- **🔧 Auto-Correction System** - Automatic step correction using multiple correction strategies
- **⚡ Core Error Detection Engine** - Real-time error detection with Gemini integration
- **💉 Prompt Injection Remediation** - Real-time error remediation with prompt injection
- **🔗 LangChain and LangGraph Interceptors** - Seamless framework integration
- **🔄 Intelligent Retry System** - Smart retry with pattern learning
- **📊 Performance Monitoring** - Comprehensive metrics and reporting
- **🛠️ CLI Interface** - Complete command-line interface with Gemini setup
- **📚 Working Examples** - Real AI integration examples and demos
## 🤝 Contributing
We welcome contributions! Please see our [Contributing Guidelines](CONTRIBUTING.md) for details.
### How to Contribute
1. **Fork** the repository
2. **Create** a feature branch (`git checkout -b feature/amazing-feature`)
3. **Commit** your changes (`git commit -m 'Add some amazing feature'`)
4. **Push** to the branch (`git push origin feature/amazing-feature`)
5. **Open** a Pull Request
### Development Guidelines
- Follow the existing code style
- Add tests for new features
- Update documentation as needed
- Ensure all tests pass
- Follow semantic commit messages
## 📄 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🆘 Support
- **Documentation**: [Full API reference](https://aigie.readthedocs.io)
- **Issues**: [Report bugs and feature requests](https://github.com/NirelNemirovsky/aigie-io/issues)
- **Discussions**: [Community discussions](https://github.com/NirelNemirovsky/aigie-io/discussions)
## 🙏 Acknowledgments
- [LangChain](https://github.com/langchain-ai/langchain) for the amazing AI framework
- [LangGraph](https://github.com/langchain-ai/langgraph) for the powerful graph-based AI workflows
- [Google Gemini](https://ai.google.dev/) for the AI analysis capabilities
---
<div align="center">
<strong>Built with ❤️ for the AI community</strong>
</div>
Raw data
{
"_id": null,
"home_page": "https://github.com/NirelNemirovsky/aigie-io",
"name": "aigie",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "Aigie Team <nirel@aigie.io>",
"keywords": "ai, artificial-intelligence, langchain, langgraph, error-detection, monitoring, validation, llm, gemini, agent, runtime, remediation, auto-correction",
"author": "Aigie Team",
"author_email": "Aigie Team <nirel@aigie.io>",
"download_url": "https://files.pythonhosted.org/packages/18/44/dc19715ba94de795d11185fb3f9a573e27d492e50546aa170d96847aeaaa/aigie-0.1.0.tar.gz",
"platform": null,
"description": "# Aigie\n\n[](https://badge.fury.io/py/aigie)\n[](https://www.python.org/downloads/)\n[](https://opensource.org/licenses/MIT)\n[](https://github.com/psf/black)\n[](https://pytest.org/)\n\n> **AI Agent Runtime Error Detection & Remediation**\n\nAigie is a real-time error detection and monitoring system for LangChain and LangGraph applications with **intelligent error remediation capabilities** and a revolutionary **LLM-as-Judge validation system**. It provides seamless integration without requiring additional code from users, automatically detecting, analyzing, validating, and fixing runtime errors as they occur.\n\n## \u2728 Features\n\n- **\ud83d\ude80 Zero-Code Integration** - Automatically detects and wraps LangChain/LangGraph applications\n- **\u26a1 Real-time Error Detection** - Immediate error reporting with classification and severity assessment\n- **\ud83e\udde0 LLM-as-Judge Validation** - Revolutionary AI-powered step validation using 6 validation strategies\n- **\ud83e\udd16 Gemini-Powered Analysis** - AI-powered error classification and intelligent remediation\n- **\ud83d\udd04 Intelligent Retry System** - Automatic retry with enhanced context from Gemini\n- **\ud83d\udc89 Prompt Injection Remediation** - Actually fixes errors by injecting guidance into AI agent prompts\n- **\ud83d\udd27 Auto-Correction System** - Automatically fixes invalid steps using multiple correction strategies\n- **\ud83d\udcca Comprehensive Monitoring** - Covers execution, API, state, and memory errors\n- **\ud83d\udcc8 Performance Insights** - Track execution time, memory usage, and resource consumption\n- **\ud83e\udde0 Pattern Learning** - Learns from successful and failed operations to improve over time\n\n## \ud83d\ude80 Quick Start\n\n### Installation\n\n```bash\npip install aigie\n```\n\n### Basic Usage (Zero Code Changes)\n\n```python\n# Just import Aigie - it automatically starts monitoring\nfrom aigie import auto_integrate\n\n# Your existing LangChain code works unchanged\nfrom langchain_core.prompts import PromptTemplate\nfrom langchain_core.runnables import RunnableSequence\n\n# Aigie automatically intercepts and monitors\nprompt = PromptTemplate.from_template(\"Tell me a joke about {topic}\")\nchain = prompt | llm\n\n# Run normally - Aigie monitors in background\nresult = chain.invoke({\"topic\": \"programming\"})\n```\n\n### LangGraph Integration\n\n```python\n# Your existing LangGraph code works unchanged\nfrom langgraph.graph import StateGraph, END\n\n# Aigie automatically monitors state transitions and node execution\ngraph = StateGraph(StateType)\n# ... your graph setup ...\napp = graph.compile()\n\n# Run normally - Aigie monitors in background\nresult = app.invoke({\"input\": \"Hello\"})\n```\n\n## \ud83e\udde0 LLM-as-Judge Validation System\n\nAigie's revolutionary **LLM-as-Judge** validation system provides real-time validation and correction of AI agent execution steps. This system ensures agents execute correctly and efficiently by continuously monitoring, validating, and automatically correcting their behavior.\n\n### \ud83c\udfaf Validation Strategies\n\nThe system uses **6 comprehensive validation strategies** to judge each execution step:\n\n| Strategy | Description |\n|----------|-------------|\n| **Goal Alignment** | Does this step advance the agent's stated goal? |\n| **Logical Consistency** | Is the step logically sound given the context? |\n| **Output Quality** | Will this likely produce appropriate output? |\n| **State Coherence** | Does this maintain consistent agent state? |\n| **Safety Compliance** | Does this follow safety guidelines? |\n| **Performance Optimality** | Is this the most efficient approach? |\n\n### \ud83d\udd27 Auto-Correction System\n\nWhen validation fails, Aigie automatically attempts correction using multiple strategies:\n\n- **Parameter Adjustment** - Fix incorrect parameters\n- **Prompt Refinement** - Improve prompts and input data\n- **Tool Substitution** - Replace wrong tools with correct ones\n- **Logic Repair** - Fix logical errors in reasoning\n- **Goal Realignment** - Align steps with agent goals\n- **State Restoration** - Fix corrupted agent state\n\n### \ud83d\udcca Advanced Features\n\n- **Parallel Validation** - Multiple strategies run simultaneously for faster processing\n- **Pattern Learning** - Learns from validation history to improve future validations\n- **Intelligent Caching** - Caches validation results for similar steps\n- **Adaptive Thresholds** - Dynamically adjusts validation criteria based on performance\n- **Rich Context Capture** - Captures comprehensive execution context for intelligent validation\n\n### Example Usage\n\n```python\nfrom aigie import auto_integrate\nfrom aigie.core.validation import ValidationEngine\n\n# Auto-integration enables LLM-as-Judge validation automatically\nauto_integrate()\n\n# Your existing code works unchanged - validation happens automatically\nfrom langchain_core.prompts import PromptTemplate\nfrom langchain_core.runnables import RunnableSequence\n\nprompt = PromptTemplate.from_template(\"Tell me a joke about {topic}\")\nchain = prompt | llm\n\n# Each step is automatically validated and corrected if needed\nresult = chain.invoke({\"topic\": \"programming\"})\n```\n\n## \ud83e\udd16 Gemini Integration\n\nAigie supports two ways to use Gemini:\n\n### 1. Vertex AI (Recommended for production)\n```bash\nexport GOOGLE_CLOUD_PROJECT=your-project-id\ngcloud auth application-default login\ngcloud services enable aiplatform.googleapis.com\n```\n\n### 2. Gemini API Key (Best for local/dev)\n```bash\nexport GEMINI_API_KEY=your-gemini-api-key\n# Get from https://aistudio.google.com/app/apikey\n```\n\n### Install Gemini dependencies\n```bash\npip install google-cloud-aiplatform vertexai google-generativeai\n```\n\n## \ud83d\udd27 Advanced Usage\n\n### Intelligent Retry with Enhanced Context\n\n```python\nfrom aigie.core.intelligent_retry import intelligent_retry\n\n@intelligent_retry(max_retries=3)\ndef my_function(input_data):\n # If this fails, Aigie will:\n # 1. Analyze the error with Gemini\n # 2. Generate enhanced retry context\n # 3. Automatically retry with better parameters\n return process_data(input_data)\n```\n\n### CLI Usage\n\n```bash\n# Enable monitoring\naigie enable --config development\n\n# Show status\naigie status\n\n# Show detailed analysis\naigie analysis\n\n# Gemini Integration\naigie gemini --setup your-project-id\naigie gemini --status\naigie gemini --test\n```\n\n## \ud83d\udccb Error Types Detected\n\n| Category | Description |\n|----------|-------------|\n| **Execution Errors** | Runtime exceptions, timeouts, infinite loops |\n| **API Errors** | External service failures, rate limits, authentication issues |\n| **State Errors** | Invalid state transitions, data corruption, type mismatches |\n| **Memory Errors** | Overflow, corruption, persistence failures |\n| **Performance Issues** | Slow execution, resource exhaustion, memory leaks |\n| **Framework-specific** | LangChain chain/tool/agent errors, LangGraph node/state errors |\n\n## \ud83d\udcca Monitoring Capabilities\n\n- **Real-time Error Logging** - Immediate error reporting with classification\n- **Performance Metrics** - Execution time, memory usage, API call latency\n- **State Tracking** - Monitor agent state changes and transitions\n- **Resource Monitoring** - CPU, memory, and disk usage with health indicators\n- **AI-Powered Analysis** - Intelligent error classification and remediation strategies\n- **Pattern Learning** - Learns from successful and failed operations\n\n## \ud83c\udfd7\ufe0f Project Structure\n\n```\naigie/\n\u251c\u2500\u2500 core/ # Core functionality\n\u2502 \u251c\u2500\u2500 types/ # Type definitions and data structures\n\u2502 \u2502 \u251c\u2500\u2500 error_types.py # Error classification and severity\n\u2502 \u2502 \u2514\u2500\u2500 validation_types.py # Validation data structures\n\u2502 \u251c\u2500\u2500 validation/ # \ud83e\udde0 LLM-as-Judge validation system\n\u2502 \u2502 \u251c\u2500\u2500 runtime_validator.py # LLM-as-Judge implementation\n\u2502 \u2502 \u251c\u2500\u2500 step_corrector.py # Auto-correction system\n\u2502 \u2502 \u251c\u2500\u2500 validation_engine.py # Main orchestrator\n\u2502 \u2502 \u251c\u2500\u2500 validation_pipeline.py # Multi-stage validation\n\u2502 \u2502 \u251c\u2500\u2500 validation_monitor.py # Performance monitoring\n\u2502 \u2502 \u2514\u2500\u2500 context_extractor.py # Context inference\n\u2502 \u251c\u2500\u2500 error_handling/ # Error detection and handling\n\u2502 \u2502 \u251c\u2500\u2500 error_detector.py # Main error detection engine\n\u2502 \u2502 \u2514\u2500\u2500 intelligent_retry.py # Smart retry system\n\u2502 \u251c\u2500\u2500 monitoring/ # Performance and resource monitoring\n\u2502 \u2502 \u2514\u2500\u2500 monitoring.py # Resource monitoring\n\u2502 \u251c\u2500\u2500 ai/ # AI/LLM components\n\u2502 \u2502 \u2514\u2500\u2500 gemini_analyzer.py # Gemini-powered analysis\n\u2502 \u2514\u2500\u2500 utils/ # Utility functions\n\u251c\u2500\u2500 interceptors/ # Framework-specific interceptors\n\u2502 \u251c\u2500\u2500 langchain.py # LangChain interceptor\n\u2502 \u251c\u2500\u2500 langgraph.py # LangGraph interceptor\n\u2502 \u2514\u2500\u2500 validation_interceptor.py # Enhanced interceptor with validation\n\u251c\u2500\u2500 reporting/ # Error reporting and logging\n\u251c\u2500\u2500 utils/ # Utility functions\n\u251c\u2500\u2500 cli.py # Command-line interface\n\u2514\u2500\u2500 auto_integration.py # Automatic integration system\n```\n\n## \u2699\ufe0f Configuration\n\n### Environment Variables\n\n```bash\nexport AIGIE_LOG_LEVEL=INFO\nexport AIGIE_ENABLE_METRICS=true\nexport AIGIE_ERROR_THRESHOLD=5\nexport AIGIE_ENABLE_ALERTS=true\n```\n\n### Configuration Files\n\n```bash\n# Generate configuration\naigie config --generate config.yml\n\n# Use configuration\naigie enable --config config.yml\n```\n\n## \ud83d\udee0\ufe0f Development\n\n### Prerequisites\n\n- Python 3.9+\n- pip\n- git\n\n### Setup Development Environment\n\n```bash\n# Clone the repository\ngit clone https://github.com/NirelNemirovsky/aigie-io.git\ncd aigie-io\n\n# Create virtual environment\npython -m venv venv\nsource venv/bin/activate # On Windows: venv\\Scripts\\activate\n\n# Install dependencies\npip install -r requirements.txt\npip install -e .\n```\n\n### Running Tests\n\n```bash\n# Run all tests\npytest tests/ -v\n\n# Run specific test categories\npytest tests/unit/ -v\npytest tests/integration/ -v\npytest tests/e2e/ -v\n```\n\n### Running Examples\n\n```bash\n# Set up Gemini (one-time)\nexport GOOGLE_CLOUD_PROJECT=your-project-id\n\n# Run comprehensive example\npython examples/ai_research_assistant.py\n```\n\n### Code Quality\n\n```bash\n# Format code\nblack aigie/ tests/ examples/\n\n# Lint code\nflake8 aigie/ tests/ examples/\n\n# Type checking\nmypy aigie/\n```\n\n## \ud83d\udcc8 Current Status\n\n\u2705 **Fully Implemented and Working**:\n- **\ud83e\udde0 LLM-as-Judge Validation System** - Revolutionary AI-powered step validation with 6 validation strategies\n- **\ud83d\udd27 Auto-Correction System** - Automatic step correction using multiple correction strategies\n- **\u26a1 Core Error Detection Engine** - Real-time error detection with Gemini integration\n- **\ud83d\udc89 Prompt Injection Remediation** - Real-time error remediation with prompt injection\n- **\ud83d\udd17 LangChain and LangGraph Interceptors** - Seamless framework integration\n- **\ud83d\udd04 Intelligent Retry System** - Smart retry with pattern learning\n- **\ud83d\udcca Performance Monitoring** - Comprehensive metrics and reporting\n- **\ud83d\udee0\ufe0f CLI Interface** - Complete command-line interface with Gemini setup\n- **\ud83d\udcda Working Examples** - Real AI integration examples and demos\n\n## \ud83e\udd1d Contributing\n\nWe welcome contributions! Please see our [Contributing Guidelines](CONTRIBUTING.md) for details.\n\n### How to Contribute\n\n1. **Fork** the repository\n2. **Create** a feature branch (`git checkout -b feature/amazing-feature`)\n3. **Commit** your changes (`git commit -m 'Add some amazing feature'`)\n4. **Push** to the branch (`git push origin feature/amazing-feature`)\n5. **Open** a Pull Request\n\n### Development Guidelines\n\n- Follow the existing code style\n- Add tests for new features\n- Update documentation as needed\n- Ensure all tests pass\n- Follow semantic commit messages\n\n## \ud83d\udcc4 License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## \ud83c\udd98 Support\n\n- **Documentation**: [Full API reference](https://aigie.readthedocs.io)\n- **Issues**: [Report bugs and feature requests](https://github.com/NirelNemirovsky/aigie-io/issues)\n- **Discussions**: [Community discussions](https://github.com/NirelNemirovsky/aigie-io/discussions)\n\n## \ud83d\ude4f Acknowledgments\n\n- [LangChain](https://github.com/langchain-ai/langchain) for the amazing AI framework\n- [LangGraph](https://github.com/langchain-ai/langgraph) for the powerful graph-based AI workflows\n- [Google Gemini](https://ai.google.dev/) for the AI analysis capabilities\n\n---\n\n<div align=\"center\">\n <strong>Built with \u2764\ufe0f for the AI community</strong>\n</div>\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "AI Agent Runtime Error Detection & Remediation with LLM-as-Judge Validation",
"version": "0.1.0",
"project_urls": {
"Bug Tracker": "https://github.com/NirelNemirovsky/aigie-io/issues",
"Documentation": "https://aigie.readthedocs.io",
"Homepage": "https://github.com/NirelNemirovsky/aigie-io",
"Repository": "https://github.com/NirelNemirovsky/aigie-io"
},
"split_keywords": [
"ai",
" artificial-intelligence",
" langchain",
" langgraph",
" error-detection",
" monitoring",
" validation",
" llm",
" gemini",
" agent",
" runtime",
" remediation",
" auto-correction"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "696766a57759e30ba90f57a45a4a151b680b00d38115ca5b7c27f3614b4f10a6",
"md5": "203f1bbb6e957ca87330725154ff9880",
"sha256": "00c43507e40cf9c90d198c3d113f4fe52c88d2428c475351d63c9a60e81c26bb"
},
"downloads": -1,
"filename": "aigie-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "203f1bbb6e957ca87330725154ff9880",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 112694,
"upload_time": "2025-09-06T19:30:59",
"upload_time_iso_8601": "2025-09-06T19:30:59.783050Z",
"url": "https://files.pythonhosted.org/packages/69/67/66a57759e30ba90f57a45a4a151b680b00d38115ca5b7c27f3614b4f10a6/aigie-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "1844dc19715ba94de795d11185fb3f9a573e27d492e50546aa170d96847aeaaa",
"md5": "537fd36fbb9640cf87dff02d05b495cd",
"sha256": "f5702272de34849c37156bf874b0c57e09cdbad3e166f08de5703fc623355f00"
},
"downloads": -1,
"filename": "aigie-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "537fd36fbb9640cf87dff02d05b495cd",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 145824,
"upload_time": "2025-09-06T19:31:01",
"upload_time_iso_8601": "2025-09-06T19:31:01.350316Z",
"url": "https://files.pythonhosted.org/packages/18/44/dc19715ba94de795d11185fb3f9a573e27d492e50546aa170d96847aeaaa/aigie-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-09-06 19:31:01",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "NirelNemirovsky",
"github_project": "aigie-io",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "langchain",
"specs": [
[
">=",
"0.1.0"
]
]
},
{
"name": "langgraph",
"specs": [
[
">=",
"0.1.0"
]
]
},
{
"name": "langchain-core",
"specs": [
[
">=",
"0.1.0"
]
]
},
{
"name": "psutil",
"specs": [
[
">=",
"5.9.0"
]
]
},
{
"name": "rich",
"specs": [
[
">=",
"13.0.0"
]
]
},
{
"name": "structlog",
"specs": [
[
">=",
"23.0.0"
]
]
},
{
"name": "pytest",
"specs": [
[
">=",
"7.0.0"
]
]
},
{
"name": "pytest-asyncio",
"specs": [
[
">=",
"0.21.0"
]
]
},
{
"name": "black",
"specs": [
[
">=",
"23.0.0"
]
]
},
{
"name": "flake8",
"specs": [
[
">=",
"6.0.0"
]
]
},
{
"name": "mypy",
"specs": [
[
">=",
"1.0.0"
]
]
},
{
"name": "google-cloud-logging",
"specs": [
[
">=",
"3.0.0"
]
]
},
{
"name": "google-cloud-monitoring",
"specs": [
[
">=",
"2.0.0"
]
]
},
{
"name": "google-generativeai",
"specs": [
[
">=",
"0.3.0"
]
]
},
{
"name": "google-cloud-aiplatform",
"specs": [
[
">=",
"1.38.0"
]
]
},
{
"name": "vertexai",
"specs": [
[
">=",
"1.38.0"
]
]
}
],
"lcname": "aigie"
}
JSON
