# Multi-Agent Debugger
A powerful Python package that uses multiple AI agents to debug API failures by analyzing logs, code, and user questions. Built with CrewAI, it supports LLM providers including OpenAI, Anthropic, Google, Ollama, and more.
## 🎥 Demo Video
Watch the multiagent-debugger in action:
[](https://youtu.be/9VTe12iVQ-A?feature=shared)
## Features
- **Multi-Agent Architecture**
- Question Analysis Agent: Extracts key entities from natural language questions
- Log Analysis Agent: Searches and filters logs for relevant information
- Code Analysis Agent: Finds API handlers, dependencies, and error handling code
- Root Cause Analysis Agent: Synthesizes findings to determine failure causes
- **Comprehensive Analysis Tools**
- Log Analysis: Search, filter, and extract stack traces from logs
- Code Analysis: Find API handlers, dependencies, and error handling patterns
- Natural Language Processing: Convert user questions into structured queries
- **Multi-Provider LLM Support**
- OpenAI (GPT-4, GPT-3.5-turbo)
- Anthropic (Claude-3 models)
- Google (Gemini models)
- Ollama (Local models)
- Azure OpenAI
- AWS Bedrock
- And 50+ more providers
- **Advanced Features**
- Dynamic Model Discovery
- Automatic Provider Configuration
- Flexible Log Path Configuration
- Customizable Code Path Analysis
- Verbose Debug Mode
## Installation
```bash
# From PyPI (coming soon)
pip install multiagent-debugger
# From source
git clone https://github.com/VishApp/multiagent-debugger.git
cd multiagent-debugger
pip install -e .
```
## Quick Start
1. Set up your configuration:
```bash
multiagent-debugger setup
```
2. Debug an API failure:
```bash
multiagent-debugger debug "Why did my /api/users endpoint fail yesterday?"
```
## Architecture
The Multi-Agent Debugger uses a sophisticated architecture that combines multiple specialized AI agents working together to analyze and debug API failures. Here's how the components interact:
### Key Components
1. **User Interface Layer**
- CLI Interface for user interaction
- Handles command processing and output formatting
2. **Configuration Layer**
- LLM Settings supporting 59+ providers
- Config Manager for handling paths and settings
3. **Debugger Crew**
- Crew Orchestrator managing agent interactions
- Coordinates the analysis workflow
4. **Specialized Agents**
- Question Analyzer: Processes natural language queries
- Log Analyzer: Examines log files
- Code Analyzer: Reviews codebase
- Root Cause Synthesizer: Combines findings
5. **Tools Layer**
- Log Tools for log file analysis
- Code Tools for codebase examination
6. **External Resources**
- Log Files
- Codebase
- LLM Providers (OpenAI, Anthropic, etc.)
## Configuration
Create a `config.yaml` file (or use the setup command):
```yaml
# Paths to log files
log_paths:
- /var/log/myapp/app.log
- /var/log/nginx/access.log
# Path to codebase
code_path: /path/to/your/code
# LLM configuration
llm:
provider: openai # or anthropic, google, ollama, etc.
model_name: gpt-4
temperature: 0.1
# api_key: optional, can use environment variable
```
### Environment Variables
Set the appropriate environment variable for your chosen provider:
- OpenAI: `OPENAI_API_KEY`
- Anthropic: `ANTHROPIC_API_KEY`
- Google: `GOOGLE_API_KEY`
- Azure: `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_ENDPOINT`
- AWS: `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, `AWS_REGION`
- See documentation for other providers
## Advanced Usage
### List Available Providers
```bash
multiagent-debugger list-providers
```
### List Models for a Provider
```bash
multiagent-debugger list-models openai
```
### Debug with Custom Config
```bash
multiagent-debugger debug "Question?" --config path/to/config.yaml
```
### Enable Verbose Output
```bash
multiagent-debugger debug "Question?" --verbose
```
## How It Works
1. **Question Analysis**
- Extracts key information like API routes, timestamps, and error types
- Structures the query for other agents
2. **Log Analysis**
- Searches through specified log files
- Filters relevant log entries
- Extracts stack traces and error patterns
3. **Code Analysis**
- Locates relevant API handlers
- Identifies dependencies and error handlers
- Maps the code structure
4. **Root Cause Analysis**
- Synthesizes information from other agents
- Determines the most likely cause
- Provides actionable insights
## Development
```bash
# Create virtual environment
python package_builder.py venv
# Install development dependencies
python package_builder.py install
# Run tests
python package_builder.py test
# Build distribution
python package_builder.py dist
```
## Requirements
- Python 3.8+
- Dependencies:
- crewai>=0.28.0
- pydantic>=2.0.0
- And others (see requirements.txt)
## License
MIT License - see [LICENSE](LICENSE) for details.
## Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
## Support
- GitHub Issues: [Report a bug](https://github.com/VishApp/multiagent-debugger/issues)
- Documentation: [Read more](https://github.com/VishApp/multiagent-debugger#readme)
Raw data
{
"_id": null,
"home_page": "https://github.com/VishApp/multiagent-debugger",
"name": "multiagent-debugger",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "debugger, api, multi-agent, crewai, llm, openai, anthropic, debugging",
"author": "Vishnu Prasad",
"author_email": "vishnuprasadapp@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/8a/a5/70fe64bf8e1c22c610fd891f43ae31f68b517d8f512978e4c02425affa31/multiagent_debugger-1.0.16.tar.gz",
"platform": null,
"description": "# Multi-Agent Debugger\n\nA powerful Python package that uses multiple AI agents to debug API failures by analyzing logs, code, and user questions. Built with CrewAI, it supports LLM providers including OpenAI, Anthropic, Google, Ollama, and more.\n\n## \ud83c\udfa5 Demo Video\n\nWatch the multiagent-debugger in action:\n\n[](https://youtu.be/9VTe12iVQ-A?feature=shared)\n\n## Features\n\n- **Multi-Agent Architecture**\n - Question Analysis Agent: Extracts key entities from natural language questions\n - Log Analysis Agent: Searches and filters logs for relevant information\n - Code Analysis Agent: Finds API handlers, dependencies, and error handling code\n - Root Cause Analysis Agent: Synthesizes findings to determine failure causes\n\n- **Comprehensive Analysis Tools**\n - Log Analysis: Search, filter, and extract stack traces from logs\n - Code Analysis: Find API handlers, dependencies, and error handling patterns\n - Natural Language Processing: Convert user questions into structured queries\n\n- **Multi-Provider LLM Support**\n - OpenAI (GPT-4, GPT-3.5-turbo)\n - Anthropic (Claude-3 models)\n - Google (Gemini models)\n - Ollama (Local models)\n - Azure OpenAI\n - AWS Bedrock\n - And 50+ more providers\n\n- **Advanced Features**\n - Dynamic Model Discovery\n - Automatic Provider Configuration\n - Flexible Log Path Configuration\n - Customizable Code Path Analysis\n - Verbose Debug Mode\n\n## Installation\n\n```bash\n# From PyPI (coming soon)\npip install multiagent-debugger\n\n# From source\ngit clone https://github.com/VishApp/multiagent-debugger.git\ncd multiagent-debugger\npip install -e .\n```\n\n## Quick Start\n\n1. Set up your configuration:\n```bash\nmultiagent-debugger setup\n```\n\n2. Debug an API failure:\n```bash\nmultiagent-debugger debug \"Why did my /api/users endpoint fail yesterday?\"\n```\n\n## Architecture\n\nThe Multi-Agent Debugger uses a sophisticated architecture that combines multiple specialized AI agents working together to analyze and debug API failures. Here's how the components interact:\n\n\n\n### Key Components\n\n1. **User Interface Layer**\n - CLI Interface for user interaction\n - Handles command processing and output formatting\n\n2. **Configuration Layer**\n - LLM Settings supporting 59+ providers\n - Config Manager for handling paths and settings\n\n3. **Debugger Crew**\n - Crew Orchestrator managing agent interactions\n - Coordinates the analysis workflow\n\n4. **Specialized Agents**\n - Question Analyzer: Processes natural language queries\n - Log Analyzer: Examines log files\n - Code Analyzer: Reviews codebase\n - Root Cause Synthesizer: Combines findings\n\n5. **Tools Layer**\n - Log Tools for log file analysis\n - Code Tools for codebase examination\n\n6. **External Resources**\n - Log Files\n - Codebase\n - LLM Providers (OpenAI, Anthropic, etc.)\n\n## Configuration\n\nCreate a `config.yaml` file (or use the setup command):\n\n```yaml\n# Paths to log files\nlog_paths:\n - /var/log/myapp/app.log\n - /var/log/nginx/access.log\n\n# Path to codebase\ncode_path: /path/to/your/code\n\n# LLM configuration\nllm:\n provider: openai # or anthropic, google, ollama, etc.\n model_name: gpt-4\n temperature: 0.1\n # api_key: optional, can use environment variable\n```\n\n### Environment Variables\n\nSet the appropriate environment variable for your chosen provider:\n\n- OpenAI: `OPENAI_API_KEY`\n- Anthropic: `ANTHROPIC_API_KEY`\n- Google: `GOOGLE_API_KEY`\n- Azure: `AZURE_OPENAI_API_KEY`, `AZURE_OPENAI_ENDPOINT`\n- AWS: `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, `AWS_REGION`\n- See documentation for other providers\n\n## Advanced Usage\n\n### List Available Providers\n```bash\nmultiagent-debugger list-providers\n```\n\n### List Models for a Provider\n```bash\nmultiagent-debugger list-models openai\n```\n\n### Debug with Custom Config\n```bash\nmultiagent-debugger debug \"Question?\" --config path/to/config.yaml\n```\n\n### Enable Verbose Output\n```bash\nmultiagent-debugger debug \"Question?\" --verbose\n```\n\n## How It Works\n\n1. **Question Analysis**\n - Extracts key information like API routes, timestamps, and error types\n - Structures the query for other agents\n\n2. **Log Analysis**\n - Searches through specified log files\n - Filters relevant log entries\n - Extracts stack traces and error patterns\n\n3. **Code Analysis**\n - Locates relevant API handlers\n - Identifies dependencies and error handlers\n - Maps the code structure\n\n4. **Root Cause Analysis**\n - Synthesizes information from other agents\n - Determines the most likely cause\n - Provides actionable insights\n\n## Development\n\n```bash\n# Create virtual environment\npython package_builder.py venv\n\n# Install development dependencies\npython package_builder.py install\n\n# Run tests\npython package_builder.py test\n\n# Build distribution\npython package_builder.py dist\n```\n\n## Requirements\n\n- Python 3.8+\n- Dependencies:\n - crewai>=0.28.0\n - pydantic>=2.0.0\n - And others (see requirements.txt)\n\n## License\n\nMIT License - see [LICENSE](LICENSE) for details.\n\n## Contributing\n\nContributions are welcome! Please feel free to submit a Pull Request.\n\n## Support\n\n- GitHub Issues: [Report a bug](https://github.com/VishApp/multiagent-debugger/issues)\n- Documentation: [Read more](https://github.com/VishApp/multiagent-debugger#readme)\n",
"bugtrack_url": null,
"license": null,
"summary": "A multi-agent system for debugging API failures",
"version": "1.0.16",
"project_urls": {
"Bug Tracker": "https://github.com/VishApp/multiagent-debugger/issues",
"Documentation": "https://github.com/VishApp/multiagent-debugger#readme",
"Homepage": "https://github.com/VishApp/multiagent-debugger",
"Source Code": "https://github.com/VishApp/multiagent-debugger"
},
"split_keywords": [
"debugger",
" api",
" multi-agent",
" crewai",
" llm",
" openai",
" anthropic",
" debugging"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "9fbcc90c5362c4d80f9a58c1647352429d2de555e804a8225376ef9472959d38",
"md5": "c1e090eaa2176142b0ffd6109721fafc",
"sha256": "52736594f62f6e7c8ad41c29bbb3de0e24fb5edd71a6e6840baca14219a04dd9"
},
"downloads": -1,
"filename": "multiagent_debugger-1.0.16-py3-none-any.whl",
"has_sig": false,
"md5_digest": "c1e090eaa2176142b0ffd6109721fafc",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 34862,
"upload_time": "2025-07-11T16:40:43",
"upload_time_iso_8601": "2025-07-11T16:40:43.687136Z",
"url": "https://files.pythonhosted.org/packages/9f/bc/c90c5362c4d80f9a58c1647352429d2de555e804a8225376ef9472959d38/multiagent_debugger-1.0.16-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "8aa570fe64bf8e1c22c610fd891f43ae31f68b517d8f512978e4c02425affa31",
"md5": "cc12fc18923c9e4f4cb00b3cf6226ef0",
"sha256": "d94da170c08635fa0185bbc311628c6bed066ada45bda9a10b7cee8663749396"
},
"downloads": -1,
"filename": "multiagent_debugger-1.0.16.tar.gz",
"has_sig": false,
"md5_digest": "cc12fc18923c9e4f4cb00b3cf6226ef0",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 31245,
"upload_time": "2025-07-11T16:40:44",
"upload_time_iso_8601": "2025-07-11T16:40:44.873262Z",
"url": "https://files.pythonhosted.org/packages/8a/a5/70fe64bf8e1c22c610fd891f43ae31f68b517d8f512978e4c02425affa31/multiagent_debugger-1.0.16.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-11 16:40:44",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "VishApp",
"github_project": "multiagent-debugger",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "crewai",
"specs": [
[
">=",
"0.28.0"
]
]
},
{
"name": "pydantic",
"specs": [
[
">=",
"2.0.0"
]
]
},
{
"name": "python-dotenv",
"specs": [
[
">=",
"1.0.0"
]
]
},
{
"name": "typer",
"specs": [
[
">=",
"0.9.0"
]
]
},
{
"name": "openai",
"specs": [
[
">=",
"1.0.0"
]
]
},
{
"name": "anthropic",
"specs": [
[
">=",
"0.8.0"
]
]
},
{
"name": "colorlog",
"specs": [
[
">=",
"6.7.0"
]
]
},
{
"name": "tree-sitter",
"specs": [
[
">=",
"0.20.1"
]
]
},
{
"name": "gitpython",
"specs": [
[
">=",
"3.1.31"
]
]
},
{
"name": "pyyaml",
"specs": [
[
">=",
"6.0"
]
]
},
{
"name": "rich",
"specs": [
[
">=",
"13.4.2"
]
]
},
{
"name": "requests",
"specs": [
[
">=",
"2.25.0"
]
]
}
],
"lcname": "multiagent-debugger"
}
JSON
