# AgentMesh
A production-grade multi-agent orchestration platform built on Microsoft AutoGen, enabling sophisticated agent collaboration through mesh networking with CLI and REST API interfaces.
## Features
### 🤖 Multi-Agent Orchestration
- **Sequential**: Linear workflows where agents build on each other's work
- **Round-Robin**: Collaborative discussions with turn-based participation
- **Graph-Based**: Complex workflows with conditional branching and parallel execution
- **Swarm Coordination**: Autonomous agent collaboration with self-organizing behavior
### 🔧 Flexible Communication
- Message passing with context preservation
- Agent handoff mechanisms with specialization routing
- Real-time monitoring and analytics
- Parameter tuning during execution
### 💻 Dual Interface
- **CLI Tool**: Complete command-line interface for workflow management
- **REST API**: Full-featured API with OpenAPI documentation
- **Configuration**: YAML-based workflow definitions with validation
### 🚀 Production Ready
- Real-time monitoring and metrics
- Performance optimization and caching
- Comprehensive logging and debugging
- Scalable architecture with AutoGen native integration
### 📊 Advanced Features
- Graph visualization (ASCII, Mermaid, JSON)
- Swarm analytics and handoff pattern analysis
- Runtime parameter tuning
- Quality gates and conditional routing
- Parallel execution and synchronization
## Quick Start
### Installation
```bash
# Install with Poetry (recommended)
poetry install
# Or with pip
pip install -e .
```
### Basic Usage
#### CLI Interface
```bash
# Create agents
agentmesh agent create --name "architect" --type "assistant" --model "gpt-4o"
agentmesh agent create --name "developer" --type "assistant" --model "gpt-4o"
# List agents
agentmesh agent list
# Create and run a workflow
agentmesh workflow create --config examples/code-review-workflow.yaml
agentmesh workflow run --id workflow-123 --task "Build a REST API for user management"
```
#### API Interface
```bash
# Start the API server
agentmesh server start --port 8000
# Create agent via API
curl -X POST "http://localhost:8000/api/v1/agents" \
-H "Content-Type: application/json" \
-d '{
"name": "architect",
"type": "assistant",
"model": "gpt-4o",
"system_message": "You are a software architect"
}'
```
## Orchestration Patterns & Examples
AutoGen A2A supports four distinct orchestration patterns, each optimized for different collaboration scenarios:
### 1. Sequential Orchestration
**Use Case**: Linear workflows where each agent builds on the previous agent's work.
```yaml
# examples/workflows/sequential-code-review.yaml
name: "Code Review Pipeline"
pattern: "sequential"
agents:
- name: "developer"
type: "assistant"
system_message: "You write clean, well-documented code"
- name: "reviewer"
type: "assistant"
system_message: "You review code for best practices and bugs"
- name: "architect"
type: "assistant"
system_message: "You ensure architectural compliance"
execution:
max_rounds: 3
timeout: 1800
```
```bash
# Run sequential workflow
agentmesh workflow create --config sequential-code-review.yaml
agentmesh workflow execute workflow-123 --task "Implement user authentication API"
```
### 2. Round-Robin Orchestration
**Use Case**: Collaborative discussions where agents take turns contributing ideas.
```yaml
# examples/workflows/round-robin-brainstorm.yaml
name: "Product Brainstorming Session"
pattern: "round_robin"
agents:
- name: "product_manager"
type: "assistant"
system_message: "You focus on user needs and business value"
- name: "designer"
type: "assistant"
system_message: "You consider user experience and interface design"
- name: "engineer"
type: "assistant"
system_message: "You evaluate technical feasibility"
termination:
type: "max_messages"
value: 12
execution:
max_rounds: 4
timeout: 2400
```
```bash
# Run round-robin workflow
agentmesh workflow create --config round-robin-brainstorm.yaml
agentmesh workflow execute workflow-123 --task "Design a mobile app for food delivery"
```
### 3. Graph-Based Orchestration
**Use Case**: Complex workflows with conditional branching, parallel execution, and quality gates.
```yaml
# examples/workflows/graph-product-development.yaml
name: "Product Development Pipeline"
pattern: "graph"
agents:
- name: "product_manager"
type: "assistant"
- name: "designer"
type: "assistant"
- name: "frontend_dev"
type: "assistant"
- name: "backend_dev"
type: "assistant"
- name: "qa_engineer"
type: "assistant"
graph:
nodes:
- id: "requirements"
agent: "product_manager"
description: "Define product requirements"
- id: "design"
agent: "designer"
description: "Create UI/UX designs"
- id: "frontend"
agent: "frontend_dev"
description: "Implement frontend"
- id: "backend"
agent: "backend_dev"
description: "Implement backend"
- id: "testing"
agent: "qa_engineer"
description: "Test and validate"
edges:
- id: "req_to_design"
source: "requirements"
target: "design"
type: "sequential"
- id: "design_to_frontend"
source: "design"
target: "frontend"
type: "parallel"
- id: "design_to_backend"
source: "design"
target: "backend"
type: "parallel"
- id: "frontend_to_testing"
source: "frontend"
target: "testing"
type: "synchronize"
- id: "backend_to_testing"
source: "backend"
target: "testing"
type: "synchronize"
conditions:
quality_gate:
type: "evaluation"
criteria: ["code_quality", "test_coverage"]
```
```bash
# Run graph workflow with visualization
agentmesh workflow create --config graph-product-development.yaml
agentmesh workflow visualize workflow-123 --format mermaid
agentmesh workflow execute workflow-123 --task "Build an e-commerce checkout system"
```
### 4. Swarm Coordination
**Use Case**: Autonomous agent collaboration with self-organizing behavior and dynamic handoffs.
```yaml
# examples/workflows/swarm-research-analysis.yaml
name: "Research Analysis Swarm"
pattern: "swarm"
agents:
- name: "data_collector"
type: "assistant"
system_message: "You excel at gathering research data from multiple sources"
- name: "analyst"
type: "assistant"
system_message: "You analyze data and identify patterns and insights"
- name: "statistician"
type: "assistant"
system_message: "You perform statistical analysis and validation"
- name: "report_generator"
type: "assistant"
system_message: "You create comprehensive research reports"
swarm:
participants:
- agent_id: "data_collector"
specializations: ["data_collection", "web_scraping", "research"]
handoff_targets: ["analyst", "statistician"]
participation_weight: 1.2
max_consecutive_turns: 3
- agent_id: "analyst"
specializations: ["data_analysis", "pattern_recognition", "insights"]
handoff_targets: ["statistician", "report_generator"]
participation_weight: 1.0
max_consecutive_turns: 4
- agent_id: "statistician"
specializations: ["statistics", "validation", "hypothesis_testing"]
handoff_targets: ["analyst", "report_generator"]
participation_weight: 0.8
max_consecutive_turns: 2
- agent_id: "report_generator"
specializations: ["writing", "reporting", "documentation"]
handoff_targets: ["analyst"]
participation_weight: 1.0
max_consecutive_turns: 3
termination:
max_messages: 25
timeout_seconds: 1800
handoff_config:
autonomous_threshold: 0.7
broadcast_threshold: 0.3
load_balancing: true
```
```bash
# Run swarm workflow with monitoring
agentmesh swarm create --config swarm-research-analysis.yaml
agentmesh swarm monitor --id swarm-123 --metrics all --refresh 5
agentmesh swarm tune --id swarm-123 --parameter participation_balance --value 0.8
agentmesh swarm analytics --id swarm-123 --output json
```
### Pattern Comparison
| Pattern | Best For | Complexity | Control | Autonomy |
|---------|----------|------------|---------|----------|
| **Sequential** | Linear workflows, step-by-step processes | Low | High | Low |
| **Round-Robin** | Collaborative discussions, brainstorming | Low | Medium | Low |
| **Graph** | Complex workflows, conditional logic | High | High | Medium |
| **Swarm** | Dynamic collaboration, emergent behavior | Medium | Medium | High |
### Advanced Examples
For complete examples with full configurations, see the `examples/workflows/` directory:
- **Sequential**: `sequential-document-processing.yaml`, `sequential-data-pipeline.yaml`
- **Round-Robin**: `round-robin-creative-writing.yaml`, `round-robin-problem-solving.yaml`
- **Graph**: `graph-code-review.yaml`, `graph-content-creation-pipeline.yaml`, `graph-advanced-product-development.yaml`
- **Swarm**: `swarm-product-development.yaml`, `swarm-content-creation.yaml`, `swarm-financial-analysis.yaml`
## Documentation
- [Getting Started Guide](docs/getting-started.md)
- [API Reference](docs/api-reference.md)
- [CLI Reference](docs/cli-reference.md)
- [Orchestration Patterns](docs/orchestration-patterns.md)
- [Graph Workflows](docs/graph-workflows.md)
- [Swarm Coordination](docs/swarm-coordination.md)
- [Examples](examples/)
### Pattern-Specific Guides
- **Sequential & Round-Robin**: Built-in patterns for linear and collaborative workflows
- **[Graph Workflows](docs/graph-workflows.md)**: Complex workflows with conditional branching, parallel execution, and quality gates
- **[Swarm Coordination](docs/swarm-coordination.md)**: Autonomous agent collaboration with self-organizing behavior
### CLI Quick Reference
```bash
# Workflow Management
agentmesh workflow create --config config.yaml
agentmesh workflow list --status running
agentmesh workflow execute workflow-123 --task "Your task here"
agentmesh workflow get workflow-123 --format json
# Graph Workflows
agentmesh workflow visualize workflow-123 --format mermaid
agentmesh workflow pause workflow-123
agentmesh workflow resume workflow-123
# Swarm Coordination
agentmesh swarm create --config swarm-config.yaml
agentmesh swarm monitor --id swarm-123 --metrics all
agentmesh swarm tune --id swarm-123 --parameter participation_balance --value 0.8
agentmesh swarm analytics --id swarm-123 --output json
# Agent Management
agentmesh agent create --name "agent" --type "assistant" --model "gpt-4o"
agentmesh agent list --format table
```
## Development
### Prerequisites
- Python 3.11+
- Poetry
- Redis (for message queuing)
- PostgreSQL (for persistence)
### Setup Development Environment
```bash
# Clone the repository
git clone <repository-url>
cd autogen-agent
# Install dependencies
poetry install
# Setup pre-commit hooks
poetry run pre-commit install
# Start development services
docker-compose up -d redis postgres
# Run tests
poetry run pytest
# Start development server
poetry run agentmesh server start --reload
```
### Project Structure
```
agentmesh/
├── src/
│ └── agentmesh/
│ ├── api/ # FastAPI application
│ ├── cli/ # CLI commands
│ ├── core/ # Core agent logic
│ ├── models/ # Data models
│ ├── services/ # Business logic
│ └── utils/ # Utilities
├── tests/ # Test suite
├── docs/ # Documentation
├── examples/ # Example workflows
└── deployment/ # Deployment configs
```
## Implementation Status
This project is currently under active development following the [Sprint Planning Document](SPRINT-PLANNING.md).
### ✅ Completed Sprints
- **Sprint 1**: Core Foundation & CLI Bootstrap
- **Sprint 2**: REST API Foundation & Agent Registry
- **Sprint 3**: Message Infrastructure & Communication
- **Sprint 4**: Security, Monitoring & Handoff Management
- **Sprint 5**: Sequential & Round-Robin Orchestration
- **Sprint 6**: Graph-based Workflows
- **Sprint 7**: Swarm Coordination
### 🔄 Current Focus: Sprint 8 - Performance Optimization & Caching
**System Status**: Production-ready for orchestration patterns, monitoring, and CLI/API interfaces.
### Key Features Available:
- ✅ **Multi-Agent Orchestration**: All four patterns (Sequential, Round-Robin, Graph, Swarm)
- ✅ **CLI Interface**: Complete command-line tool with workflow management
- ✅ **REST API**: Full API with OpenAPI documentation
- ✅ **Monitoring**: Real-time metrics, analytics, and visualization
- ✅ **Configuration**: YAML-based workflow definitions with validation
- ✅ **Examples**: Comprehensive examples for all orchestration patterns
See the full [implementation roadmap](SPRINT-PLANNING.md) for detailed progress tracking.
## Contributing
1. Fork the repository
2. Create a feature branch (`git checkout -b feature/amazing-feature`)
3. Commit your changes (`git commit -m 'Add amazing feature'`)
4. Push to the branch (`git push origin feature/amazing-feature`)
5. Open a Pull Request
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## Related Projects
- [Microsoft AutoGen](https://github.com/microsoft/autogen)
- [AutoGen Studio](https://github.com/microsoft/autogen/tree/main/python/packages/autogen-studio)
- [AutoGen Core](https://github.com/microsoft/autogen/tree/main/python/packages/autogen-core)
Raw data
{
"_id": null,
"home_page": "https://github.com/akhilthomas236/agentmesh",
"name": "agentmesh-orchestrator",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.14,>=3.11",
"maintainer_email": null,
"keywords": "multi-agent, orchestration, autogen, ai, workflow",
"author": "Development Team",
"author_email": "dev@agentmesh.ai",
"download_url": "https://files.pythonhosted.org/packages/9b/b9/ae8ba7988d78b1765b63b687ce6239faa82a48013309fa8ca62fd9fe0e5e/agentmesh_orchestrator-0.1.7.tar.gz",
"platform": null,
"description": "# AgentMesh\n\nA production-grade multi-agent orchestration platform built on Microsoft AutoGen, enabling sophisticated agent collaboration through mesh networking with CLI and REST API interfaces.\n\n## Features\n\n### \ud83e\udd16 Multi-Agent Orchestration\n- **Sequential**: Linear workflows where agents build on each other's work\n- **Round-Robin**: Collaborative discussions with turn-based participation\n- **Graph-Based**: Complex workflows with conditional branching and parallel execution\n- **Swarm Coordination**: Autonomous agent collaboration with self-organizing behavior\n\n### \ud83d\udd27 Flexible Communication\n- Message passing with context preservation\n- Agent handoff mechanisms with specialization routing\n- Real-time monitoring and analytics\n- Parameter tuning during execution\n\n### \ud83d\udcbb Dual Interface\n- **CLI Tool**: Complete command-line interface for workflow management\n- **REST API**: Full-featured API with OpenAPI documentation\n- **Configuration**: YAML-based workflow definitions with validation\n\n### \ud83d\ude80 Production Ready\n- Real-time monitoring and metrics\n- Performance optimization and caching\n- Comprehensive logging and debugging\n- Scalable architecture with AutoGen native integration\n\n### \ud83d\udcca Advanced Features\n- Graph visualization (ASCII, Mermaid, JSON)\n- Swarm analytics and handoff pattern analysis\n- Runtime parameter tuning\n- Quality gates and conditional routing\n- Parallel execution and synchronization\n\n## Quick Start\n\n### Installation\n\n```bash\n# Install with Poetry (recommended)\npoetry install\n\n# Or with pip\npip install -e .\n```\n\n### Basic Usage\n\n#### CLI Interface\n\n```bash\n# Create agents\nagentmesh agent create --name \"architect\" --type \"assistant\" --model \"gpt-4o\"\nagentmesh agent create --name \"developer\" --type \"assistant\" --model \"gpt-4o\"\n\n# List agents\nagentmesh agent list\n\n# Create and run a workflow\nagentmesh workflow create --config examples/code-review-workflow.yaml\nagentmesh workflow run --id workflow-123 --task \"Build a REST API for user management\"\n```\n\n#### API Interface\n\n```bash\n# Start the API server\nagentmesh server start --port 8000\n\n# Create agent via API\ncurl -X POST \"http://localhost:8000/api/v1/agents\" \\\n -H \"Content-Type: application/json\" \\\n -d '{\n \"name\": \"architect\",\n \"type\": \"assistant\",\n \"model\": \"gpt-4o\",\n \"system_message\": \"You are a software architect\"\n }'\n```\n\n## Orchestration Patterns & Examples\n\nAutoGen A2A supports four distinct orchestration patterns, each optimized for different collaboration scenarios:\n\n### 1. Sequential Orchestration\n**Use Case**: Linear workflows where each agent builds on the previous agent's work.\n\n```yaml\n# examples/workflows/sequential-code-review.yaml\nname: \"Code Review Pipeline\"\npattern: \"sequential\"\n\nagents:\n - name: \"developer\"\n type: \"assistant\"\n system_message: \"You write clean, well-documented code\"\n - name: \"reviewer\"\n type: \"assistant\"\n system_message: \"You review code for best practices and bugs\"\n - name: \"architect\"\n type: \"assistant\"\n system_message: \"You ensure architectural compliance\"\n\nexecution:\n max_rounds: 3\n timeout: 1800\n```\n\n```bash\n# Run sequential workflow\nagentmesh workflow create --config sequential-code-review.yaml\nagentmesh workflow execute workflow-123 --task \"Implement user authentication API\"\n```\n\n### 2. Round-Robin Orchestration\n**Use Case**: Collaborative discussions where agents take turns contributing ideas.\n\n```yaml\n# examples/workflows/round-robin-brainstorm.yaml\nname: \"Product Brainstorming Session\"\npattern: \"round_robin\"\n\nagents:\n - name: \"product_manager\"\n type: \"assistant\"\n system_message: \"You focus on user needs and business value\"\n - name: \"designer\"\n type: \"assistant\"\n system_message: \"You consider user experience and interface design\"\n - name: \"engineer\"\n type: \"assistant\"\n system_message: \"You evaluate technical feasibility\"\n\ntermination:\n type: \"max_messages\"\n value: 12\n\nexecution:\n max_rounds: 4\n timeout: 2400\n```\n\n```bash\n# Run round-robin workflow\nagentmesh workflow create --config round-robin-brainstorm.yaml\nagentmesh workflow execute workflow-123 --task \"Design a mobile app for food delivery\"\n```\n\n### 3. Graph-Based Orchestration\n**Use Case**: Complex workflows with conditional branching, parallel execution, and quality gates.\n\n```yaml\n# examples/workflows/graph-product-development.yaml\nname: \"Product Development Pipeline\"\npattern: \"graph\"\n\nagents:\n - name: \"product_manager\"\n type: \"assistant\"\n - name: \"designer\"\n type: \"assistant\"\n - name: \"frontend_dev\"\n type: \"assistant\"\n - name: \"backend_dev\"\n type: \"assistant\"\n - name: \"qa_engineer\"\n type: \"assistant\"\n\ngraph:\n nodes:\n - id: \"requirements\"\n agent: \"product_manager\"\n description: \"Define product requirements\"\n - id: \"design\"\n agent: \"designer\"\n description: \"Create UI/UX designs\"\n - id: \"frontend\"\n agent: \"frontend_dev\"\n description: \"Implement frontend\"\n - id: \"backend\"\n agent: \"backend_dev\"\n description: \"Implement backend\"\n - id: \"testing\"\n agent: \"qa_engineer\"\n description: \"Test and validate\"\n \n edges:\n - id: \"req_to_design\"\n source: \"requirements\"\n target: \"design\"\n type: \"sequential\"\n - id: \"design_to_frontend\"\n source: \"design\"\n target: \"frontend\"\n type: \"parallel\"\n - id: \"design_to_backend\"\n source: \"design\"\n target: \"backend\"\n type: \"parallel\"\n - id: \"frontend_to_testing\"\n source: \"frontend\"\n target: \"testing\"\n type: \"synchronize\"\n - id: \"backend_to_testing\"\n source: \"backend\"\n target: \"testing\"\n type: \"synchronize\"\n \n conditions:\n quality_gate:\n type: \"evaluation\"\n criteria: [\"code_quality\", \"test_coverage\"]\n```\n\n```bash\n# Run graph workflow with visualization\nagentmesh workflow create --config graph-product-development.yaml\nagentmesh workflow visualize workflow-123 --format mermaid\nagentmesh workflow execute workflow-123 --task \"Build an e-commerce checkout system\"\n```\n\n### 4. Swarm Coordination\n**Use Case**: Autonomous agent collaboration with self-organizing behavior and dynamic handoffs.\n\n```yaml\n# examples/workflows/swarm-research-analysis.yaml\nname: \"Research Analysis Swarm\"\npattern: \"swarm\"\n\nagents:\n - name: \"data_collector\"\n type: \"assistant\"\n system_message: \"You excel at gathering research data from multiple sources\"\n - name: \"analyst\"\n type: \"assistant\"\n system_message: \"You analyze data and identify patterns and insights\"\n - name: \"statistician\"\n type: \"assistant\"\n system_message: \"You perform statistical analysis and validation\"\n - name: \"report_generator\"\n type: \"assistant\"\n system_message: \"You create comprehensive research reports\"\n\nswarm:\n participants:\n - agent_id: \"data_collector\"\n specializations: [\"data_collection\", \"web_scraping\", \"research\"]\n handoff_targets: [\"analyst\", \"statistician\"]\n participation_weight: 1.2\n max_consecutive_turns: 3\n \n - agent_id: \"analyst\"\n specializations: [\"data_analysis\", \"pattern_recognition\", \"insights\"]\n handoff_targets: [\"statistician\", \"report_generator\"]\n participation_weight: 1.0\n max_consecutive_turns: 4\n \n - agent_id: \"statistician\"\n specializations: [\"statistics\", \"validation\", \"hypothesis_testing\"]\n handoff_targets: [\"analyst\", \"report_generator\"]\n participation_weight: 0.8\n max_consecutive_turns: 2\n \n - agent_id: \"report_generator\"\n specializations: [\"writing\", \"reporting\", \"documentation\"]\n handoff_targets: [\"analyst\"]\n participation_weight: 1.0\n max_consecutive_turns: 3\n\n termination:\n max_messages: 25\n timeout_seconds: 1800\n\n handoff_config:\n autonomous_threshold: 0.7\n broadcast_threshold: 0.3\n load_balancing: true\n```\n\n```bash\n# Run swarm workflow with monitoring\nagentmesh swarm create --config swarm-research-analysis.yaml\nagentmesh swarm monitor --id swarm-123 --metrics all --refresh 5\nagentmesh swarm tune --id swarm-123 --parameter participation_balance --value 0.8\nagentmesh swarm analytics --id swarm-123 --output json\n```\n\n### Pattern Comparison\n\n| Pattern | Best For | Complexity | Control | Autonomy |\n|---------|----------|------------|---------|----------|\n| **Sequential** | Linear workflows, step-by-step processes | Low | High | Low |\n| **Round-Robin** | Collaborative discussions, brainstorming | Low | Medium | Low |\n| **Graph** | Complex workflows, conditional logic | High | High | Medium |\n| **Swarm** | Dynamic collaboration, emergent behavior | Medium | Medium | High |\n\n### Advanced Examples\n\nFor complete examples with full configurations, see the `examples/workflows/` directory:\n\n- **Sequential**: `sequential-document-processing.yaml`, `sequential-data-pipeline.yaml`\n- **Round-Robin**: `round-robin-creative-writing.yaml`, `round-robin-problem-solving.yaml`\n- **Graph**: `graph-code-review.yaml`, `graph-content-creation-pipeline.yaml`, `graph-advanced-product-development.yaml`\n- **Swarm**: `swarm-product-development.yaml`, `swarm-content-creation.yaml`, `swarm-financial-analysis.yaml`\n\n## Documentation\n\n- [Getting Started Guide](docs/getting-started.md)\n- [API Reference](docs/api-reference.md)\n- [CLI Reference](docs/cli-reference.md)\n- [Orchestration Patterns](docs/orchestration-patterns.md)\n- [Graph Workflows](docs/graph-workflows.md)\n- [Swarm Coordination](docs/swarm-coordination.md)\n- [Examples](examples/)\n\n### Pattern-Specific Guides\n\n- **Sequential & Round-Robin**: Built-in patterns for linear and collaborative workflows\n- **[Graph Workflows](docs/graph-workflows.md)**: Complex workflows with conditional branching, parallel execution, and quality gates\n- **[Swarm Coordination](docs/swarm-coordination.md)**: Autonomous agent collaboration with self-organizing behavior\n\n### CLI Quick Reference\n\n```bash\n# Workflow Management\nagentmesh workflow create --config config.yaml\nagentmesh workflow list --status running\nagentmesh workflow execute workflow-123 --task \"Your task here\"\nagentmesh workflow get workflow-123 --format json\n\n# Graph Workflows\nagentmesh workflow visualize workflow-123 --format mermaid\nagentmesh workflow pause workflow-123\nagentmesh workflow resume workflow-123\n\n# Swarm Coordination\nagentmesh swarm create --config swarm-config.yaml\nagentmesh swarm monitor --id swarm-123 --metrics all\nagentmesh swarm tune --id swarm-123 --parameter participation_balance --value 0.8\nagentmesh swarm analytics --id swarm-123 --output json\n\n# Agent Management\nagentmesh agent create --name \"agent\" --type \"assistant\" --model \"gpt-4o\"\nagentmesh agent list --format table\n```\n\n## Development\n\n### Prerequisites\n\n- Python 3.11+\n- Poetry\n- Redis (for message queuing)\n- PostgreSQL (for persistence)\n\n### Setup Development Environment\n\n```bash\n# Clone the repository\ngit clone <repository-url>\ncd autogen-agent\n\n# Install dependencies\npoetry install\n\n# Setup pre-commit hooks\npoetry run pre-commit install\n\n# Start development services\ndocker-compose up -d redis postgres\n\n# Run tests\npoetry run pytest\n\n# Start development server\npoetry run agentmesh server start --reload\n```\n\n### Project Structure\n\n```\nagentmesh/\n\u251c\u2500\u2500 src/\n\u2502 \u2514\u2500\u2500 agentmesh/\n\u2502 \u251c\u2500\u2500 api/ # FastAPI application\n\u2502 \u251c\u2500\u2500 cli/ # CLI commands\n\u2502 \u251c\u2500\u2500 core/ # Core agent logic\n\u2502 \u251c\u2500\u2500 models/ # Data models\n\u2502 \u251c\u2500\u2500 services/ # Business logic\n\u2502 \u2514\u2500\u2500 utils/ # Utilities\n\u251c\u2500\u2500 tests/ # Test suite\n\u251c\u2500\u2500 docs/ # Documentation\n\u251c\u2500\u2500 examples/ # Example workflows\n\u2514\u2500\u2500 deployment/ # Deployment configs\n```\n\n## Implementation Status\n\nThis project is currently under active development following the [Sprint Planning Document](SPRINT-PLANNING.md).\n\n### \u2705 Completed Sprints\n\n- **Sprint 1**: Core Foundation & CLI Bootstrap \n- **Sprint 2**: REST API Foundation & Agent Registry\n- **Sprint 3**: Message Infrastructure & Communication\n- **Sprint 4**: Security, Monitoring & Handoff Management\n- **Sprint 5**: Sequential & Round-Robin Orchestration\n- **Sprint 6**: Graph-based Workflows\n- **Sprint 7**: Swarm Coordination\n\n### \ud83d\udd04 Current Focus: Sprint 8 - Performance Optimization & Caching\n\n**System Status**: Production-ready for orchestration patterns, monitoring, and CLI/API interfaces.\n\n### Key Features Available:\n\n- \u2705 **Multi-Agent Orchestration**: All four patterns (Sequential, Round-Robin, Graph, Swarm)\n- \u2705 **CLI Interface**: Complete command-line tool with workflow management\n- \u2705 **REST API**: Full API with OpenAPI documentation\n- \u2705 **Monitoring**: Real-time metrics, analytics, and visualization\n- \u2705 **Configuration**: YAML-based workflow definitions with validation\n- \u2705 **Examples**: Comprehensive examples for all orchestration patterns\n\nSee the full [implementation roadmap](SPRINT-PLANNING.md) for detailed progress tracking.\n\n## Contributing\n\n1. Fork the repository\n2. Create a feature branch (`git checkout -b feature/amazing-feature`)\n3. Commit your changes (`git commit -m 'Add amazing feature'`)\n4. Push to the branch (`git push origin feature/amazing-feature`)\n5. Open a Pull Request\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## Related Projects\n\n- [Microsoft AutoGen](https://github.com/microsoft/autogen)\n- [AutoGen Studio](https://github.com/microsoft/autogen/tree/main/python/packages/autogen-studio)\n- [AutoGen Core](https://github.com/microsoft/autogen/tree/main/python/packages/autogen-core)\n",
"bugtrack_url": null,
"license": null,
"summary": "AgentMesh - Multi-Agent Orchestration Platform",
"version": "0.1.7",
"project_urls": {
"Documentation": "https://github.com/akhilthomas236/agentmesh/blob/main/README.md",
"Homepage": "https://github.com/akhilthomas236/agentmesh",
"Repository": "https://github.com/akhilthomas236/agentmesh"
},
"split_keywords": [
"multi-agent",
" orchestration",
" autogen",
" ai",
" workflow"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d2a37548edf56b3295788c50d6a062b9394ff61c44d5518cfc736ad94d583d40",
"md5": "9e9d3d65fda395fa44888e13c6e8863b",
"sha256": "eb9d4b27663908eec34d3cda3b736401769298449eb9020a8426bcea93a772d5"
},
"downloads": -1,
"filename": "agentmesh_orchestrator-0.1.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "9e9d3d65fda395fa44888e13c6e8863b",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.14,>=3.11",
"size": 111887,
"upload_time": "2025-08-15T04:43:25",
"upload_time_iso_8601": "2025-08-15T04:43:25.230362Z",
"url": "https://files.pythonhosted.org/packages/d2/a3/7548edf56b3295788c50d6a062b9394ff61c44d5518cfc736ad94d583d40/agentmesh_orchestrator-0.1.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "9bb9ae8ba7988d78b1765b63b687ce6239faa82a48013309fa8ca62fd9fe0e5e",
"md5": "965433c4499f6fbdac643990916cefd1",
"sha256": "28aaccd89c62bd2591110eeaf77297c0ffbf74d711233a073964a43a4bb9e220"
},
"downloads": -1,
"filename": "agentmesh_orchestrator-0.1.7.tar.gz",
"has_sig": false,
"md5_digest": "965433c4499f6fbdac643990916cefd1",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.14,>=3.11",
"size": 90044,
"upload_time": "2025-08-15T04:43:27",
"upload_time_iso_8601": "2025-08-15T04:43:27.338228Z",
"url": "https://files.pythonhosted.org/packages/9b/b9/ae8ba7988d78b1765b63b687ce6239faa82a48013309fa8ca62fd9fe0e5e/agentmesh_orchestrator-0.1.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-08-15 04:43:27",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "akhilthomas236",
"github_project": "agentmesh",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "agentmesh-orchestrator"
}
JSON
