# Agent Expert Panel - Core Package
[](https://codecov.io/gh/zbloss/agent-expert-panel)
## Overview
This is the core package of the Agent Expert Panel monorepo, containing the main Python library for multi-agent AI collaboration. The framework orchestrates AI experts to solve complex problems through collaborative reasoning.
## Features
- **5 Specialized AI Experts**: Advocate, Critic, Pragmatist, Research Specialist, and Innovator
- **GraphRAG-Enhanced Memory**: Persistent knowledge across sessions with intelligent indexing
- **Real-Time Web Research**: Tavily-powered search with automatic synthesis
- **Interactive CLI**: Built with Typer for modern command-line experience
- **Human Participation**: Join AI experts in discussions via UserProxyAgent
- **Flexible Discussion Patterns**: Round-robin, structured debate, and more
## Installation
From the monorepo root:
```bash
# Install core package dependencies
cd packages/core
uv sync --group dev --group test
# Or using pip
pip install -e ".[dev]"
```
## Usage
### CLI Commands
```bash
# Interactive mode
agent-panel
# Run specific discussion
agent-panel discuss "Should we adopt microservices?" --pattern round-robin --rounds 3
# Virtual expert panel with research
agent-panel virtual-solve "How to improve customer retention?" --domain business
```
### Programmatic Usage
```python
import asyncio
from agent_expert_panel.panel import ExpertPanel, DiscussionPattern
async def main():
panel = ExpertPanel()
result = await panel.discuss(
topic="How can we improve team productivity?",
pattern=DiscussionPattern.ROUND_ROBIN,
max_rounds=3
)
print(f"Recommendation: {result.final_recommendation}")
asyncio.run(main())
```
## Configuration
Agents are configured via YAML files in the `configs/` directory. Each agent has customizable:
- Model parameters
- System messages
- API endpoints
- Timeout settings
## Development
```bash
# Run tests
pytest tests
# Format code
ruff format
# Lint code
ruff check --fix
```
## Architecture
The core package contains:
- `/src/agent_expert_panel/` - Main package source
- `/tests/` - Test suite with pytest
- `/examples/` - Usage examples
- `/docs/` - Documentation
- `/configs/` - Agent configuration files
For complete documentation, see the [main repository README](../../README.md).
Raw data
{
"_id": null,
"home_page": null,
"name": "agent-expert-panel",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.12",
"maintainer_email": null,
"keywords": "ai, agents, expert-panel, multi-agent, discussion",
"author": null,
"author_email": "Zachary Bloss <zacharybloss@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/a3/5a/4ccf315182f5df6ec2cdfbcb52e352a05f4bf0c866645543574f2c3fb893/agent_expert_panel-0.5.2.tar.gz",
"platform": null,
"description": "# Agent Expert Panel - Core Package\n\n[](https://codecov.io/gh/zbloss/agent-expert-panel)\n\n## Overview\n\nThis is the core package of the Agent Expert Panel monorepo, containing the main Python library for multi-agent AI collaboration. The framework orchestrates AI experts to solve complex problems through collaborative reasoning.\n\n## Features\n\n- **5 Specialized AI Experts**: Advocate, Critic, Pragmatist, Research Specialist, and Innovator\n- **GraphRAG-Enhanced Memory**: Persistent knowledge across sessions with intelligent indexing\n- **Real-Time Web Research**: Tavily-powered search with automatic synthesis\n- **Interactive CLI**: Built with Typer for modern command-line experience\n- **Human Participation**: Join AI experts in discussions via UserProxyAgent\n- **Flexible Discussion Patterns**: Round-robin, structured debate, and more\n\n## Installation\n\nFrom the monorepo root:\n\n```bash\n# Install core package dependencies\ncd packages/core\nuv sync --group dev --group test\n\n# Or using pip\npip install -e \".[dev]\"\n```\n\n## Usage\n\n### CLI Commands\n\n```bash\n# Interactive mode\nagent-panel\n\n# Run specific discussion\nagent-panel discuss \"Should we adopt microservices?\" --pattern round-robin --rounds 3\n\n# Virtual expert panel with research\nagent-panel virtual-solve \"How to improve customer retention?\" --domain business\n```\n\n### Programmatic Usage\n\n```python\nimport asyncio\nfrom agent_expert_panel.panel import ExpertPanel, DiscussionPattern\n\nasync def main():\n panel = ExpertPanel()\n result = await panel.discuss(\n topic=\"How can we improve team productivity?\",\n pattern=DiscussionPattern.ROUND_ROBIN,\n max_rounds=3\n )\n print(f\"Recommendation: {result.final_recommendation}\")\n\nasyncio.run(main())\n```\n\n## Configuration\n\nAgents are configured via YAML files in the `configs/` directory. Each agent has customizable:\n- Model parameters\n- System messages\n- API endpoints\n- Timeout settings\n\n## Development\n\n```bash\n# Run tests\npytest tests\n\n# Format code\nruff format\n\n# Lint code\nruff check --fix\n```\n\n## Architecture\n\nThe core package contains:\n- `/src/agent_expert_panel/` - Main package source\n- `/tests/` - Test suite with pytest\n- `/examples/` - Usage examples\n- `/docs/` - Documentation\n- `/configs/` - Agent configuration files\n\nFor complete documentation, see the [main repository README](../../README.md).\n",
"bugtrack_url": null,
"license": null,
"summary": "A panel of experts that will solve problems and discuss topics",
"version": "0.5.2",
"project_urls": {
"Documentation": "https://github.com/zbloss/agent-expert-panel#readme",
"Homepage": "https://github.com/zbloss/agent-expert-panel",
"Issues": "https://github.com/zbloss/agent-expert-panel/issues",
"Repository": "https://github.com/zbloss/agent-expert-panel"
},
"split_keywords": [
"ai",
" agents",
" expert-panel",
" multi-agent",
" discussion"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "ded12753afd476d7f53ff1ff192b83710ccd2bdee559f0996826307b7845d559",
"md5": "dc828cfca133605b003419fb0af2dd86",
"sha256": "d8e29db836b38424532d0eae3184131313e2ff785a5d51b1c5023e4b9f21b9c4"
},
"downloads": -1,
"filename": "agent_expert_panel-0.5.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "dc828cfca133605b003419fb0af2dd86",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.12",
"size": 211527,
"upload_time": "2025-09-06T02:07:28",
"upload_time_iso_8601": "2025-09-06T02:07:28.753669Z",
"url": "https://files.pythonhosted.org/packages/de/d1/2753afd476d7f53ff1ff192b83710ccd2bdee559f0996826307b7845d559/agent_expert_panel-0.5.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "a35a4ccf315182f5df6ec2cdfbcb52e352a05f4bf0c866645543574f2c3fb893",
"md5": "41e67b44cb39ee4d8a3dfd6acf3b72a9",
"sha256": "ca0615cc5967670e4eea6de2accf871d8e7f4a48c898eb1793bc129c64b80cc0"
},
"downloads": -1,
"filename": "agent_expert_panel-0.5.2.tar.gz",
"has_sig": false,
"md5_digest": "41e67b44cb39ee4d8a3dfd6acf3b72a9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.12",
"size": 225053,
"upload_time": "2025-09-06T02:07:30",
"upload_time_iso_8601": "2025-09-06T02:07:30.338499Z",
"url": "https://files.pythonhosted.org/packages/a3/5a/4ccf315182f5df6ec2cdfbcb52e352a05f4bf0c866645543574f2c3fb893/agent_expert_panel-0.5.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-09-06 02:07:30",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "zbloss",
"github_project": "agent-expert-panel#readme",
"github_not_found": true,
"lcname": "agent-expert-panel"
}
JSON
