Skip to main content

mia-anemoi: Agent-to-Agent Communication System

mia-anemoi is Mia's fork of Coral-Protocol/Anemoi — a semi-centralized multi-agent system (MAS) built on Agent-to-Agent (A2A) communication via MCP.

This fork is adapted for terminal-based agent orchestration and SimExp integration.

Anemoi Concept


🚀 Key Features

  • Semi-Centralized Architecture: Reduces dependency on a single planner agent
  • Direct Agent-to-Agent Collaboration: Real-time monitoring, assessment, and refinement
  • Efficient Context Management: Minimizes redundant prompt concatenation
  • Benchmark Performance: 52.73% accuracy on GAIA benchmark (+9.09% over OWL)
  • SimExp Integration: Terminal-to-AI session continuations with A2A communication

📦 Package Components

Component Description Location
Python Client Lightweight A2A communication library mia_anemoi/
CAMEL Agents Specialized AI agents (planning, web, coding) Anemoi/agents/
Kotlin MCP Server SSE-based thread messaging src/main/kotlin/

🔧 Installation

Python Client (Lightweight)

pip install mia-anemoi

With CAMEL Agents (Full)

pip install mia-anemoi[agents]

From Source

git clone https://github.com/miadisabelle/mia-anemoi.git
cd mia-anemoi
pip install -e .

🚀 Quick Start

1. Start MCP Server

# Using Docker
docker run -p 5555:5555 ghcr.io/miadisabelle/mia-anemoi-server:latest

# Or build and run locally
./gradlew build
java -jar build/libs/mia-anemoi-server-1.0-SNAPSHOT.jar

2. Use Python Client

from mia_anemoi import AnemoiClient, AnemoiClientConfig

# Initialize client
config = AnemoiClientConfig(transport="file", base_path="~/.anemoi")
client = AnemoiClient("session-123", config)

# Fork session with context inheritance
child_id = client.fork_session(parent_state)

# Send update to siblings
client.send_update("Task completed", {"files": ["main.py"]})

# Wait for messages
msg = client.wait_for_mentions(timeout=30)

3. Run CAMEL Agents

export CORAL_CONNECTION_URL="http://localhost:5555/devmode/gaia/public/session1/sse"
export OPENAI_API_KEY="sk-..."

python -m Anemoi.agents.planning_agent

🏗️ Architecture

┌─────────────────────────────────────────────────────────────────────┐
│                         mia-anemoi Package                           │
├─────────────────────────────────────────────────────────────────────┤
│                                                                      │
│  ┌────────────────────┐      ┌────────────────────────────────────┐ │
│  │    Python Client   │ SSE  │         Kotlin MCP Server          │ │
│  │   (mia_anemoi/)    │──────│   (src/main/kotlin/coralserver/)   │ │
│  │                    │      │                                    │ │
│  │  - AnemoiClient    │      │  - list_agents                     │ │
│  │  - FileTransport   │      │  - create_thread                   │ │
│  │  - MCPTransport    │      │  - send_message                    │ │
│  └────────────────────┘      │  - wait_for_mentions               │ │
│           │                  │  - close_thread                    │ │
│           │                  └────────────────────────────────────┘ │
│           ▼                                                         │
│  ┌────────────────────┐                                             │
│  │   CAMEL Agents     │                                             │
│  │   (Anemoi/agents/) │                                             │
│  │                    │                                             │
│  │  - planning_agent  │                                             │
│  │  - web_agent       │                                             │
│  │  - coding_agent    │                                             │
│  │  - critique_agent  │                                             │
│  └────────────────────┘                                             │
└─────────────────────────────────────────────────────────────────────┘

📄 Documentation

See the rispecs/ directory for RISE specifications:


📄 Original Publication

This is a fork of the original Anemoi project. Please cite the original work:

@article{ren2025anemoi,
  title={Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol},
  author={Ren, Xinxing and Forder, Caelum and Zang, Qianbo and Tahir, Ahsen and Georgio, Roman J. and Deb, Suman and Carroll, Peter and G\"{u}rcan, \"{O}nder and Guo, Zekun},
  journal={arXiv preprint arXiv:2508.17068},
  year={2025},
  url={https://arxiv.org/abs/2508.17068}
}


📜 License

MIT License - See LICENSE for details.

Metadata

Release files for mia-anemoi 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for mia-anemoi 0.1.0
File Size Uploaded Provenance
mia_anemoi-0.1.0.tar.gz 14.5 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for mia-anemoi 0.1.0
File Interpreter ABI Platform
mia_anemoi-0.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 27.5 kB

Release files / mia_anemoi-0.1.0.tar.gz

Download URL mia_anemoi-0.1.0.tar.gz
Size 14.5 kB
Tags Source
SHA-256 checksum
How to use checksums
022057ac7ea82bc3fc229934f459b31409cd4de46004d28394675fe2133c6635
BLAKE2b-256 checksum
How to use checksums
94d57c254ca7e83d17716ffc4625dea17650ba0e090d58bdaa4a1aa4b9a28c14
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.4

Release files / mia_anemoi-0.1.0-py3-none-any.whl

Download URL mia_anemoi-0.1.0-py3-none-any.whl
Size 13.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
0ccdd589ab1965da87595face761dd676050958c08050ac8045b525d0aab13f0
BLAKE2b-256 checksum
How to use checksums
4d37fc71151c0a70c3c34dae82b889b8831c5e84d28c7e0a301cff9fa325e52d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.4

Security for mia-anemoi 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page