Skip to content
Back to search

Anemoi

Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/coral-protocol--anemoi/badge.svg)](/listings/coral-protocol--anemoi)

HTML

<a href="/listings/coral-protocol--anemoi"><img src="/api/listings/coral-protocol--anemoi/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 96 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol
  2. **Anemoi** is a semi-centralized multi-agent system (MAS) built on a **Agent-to-Agent (A2A) communication MCP server**.
  3. Unlike traditional **context-engineering + centralized** paradigms, Anemoi introduces **structured, direct inter-agent communication** — enabling agents to collaborate much like a real-world team.
  4. 🌀 *Like winds connecting distant lands, Anemoi enables agents to communicate directly in a semi-centralized network, achieving scalable coordination and seamless information flow.*
  5. <p align="center">
  6. <img src="Anemoi/images/Anemoi_semi.png" alt="Anemoi Concept" width="70%">
  7. </p>
  8. ---
  9. ## 🚀 Key Features
  10. * **Semi-Centralized Architecture**
  11. Reduces dependency on a single planner agent, supporting adaptive plan updates.
  12. * **Direct Agent-to-Agent Collaboration**
  13. Agents can monitor progress, assess results, identify bottlenecks, and propose refinements in real time.
  14. * **Efficient Context Management**
  15. Minimizes redundant prompt concatenation and information loss, improving scalability and cost-efficiency.
  16. * **Benchmark Performance**
  17. Achieved **52.73% accuracy** on the validation set of the GAIA benchmark, setting the **state-of-the-art** among small-LLM-based systems.
  18. > Surpasses OWL (43.63%) by **+9.09%** in the same worker agents and models configuration (gpt-4.1-mini as planner agent/ gpt-4o as worker agent).
  19. <p align="center">
  20. <img src="Anemoi/images/Anemoi_workflow.png" alt="Anemoi Workflow" width="85%">
  21. </p>
  22. ---
  23. ## 📄 Publication
  24. Our work has been released on arXiv:
  25. 👉 [Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol](https://arxiv.org/abs/2508.17068)
  26. If you find this project useful, please consider citing our paper:
  27. ```
  28. @article{ren2025anemoi,
  29. title={Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol},
  30. author={Ren, Xinxing and Forder, Caelum and Zang, Qianbo and Tahir, Ahsen and Georgio, Roman J. and Deb, Suman and Carroll, Peter and Gürcan, Önder and Guo, Zekun},
  31. journal={arXiv preprint arXiv:2508.17068},
  32. year={2025},
  33. url={https://arxiv.org/abs/2508.17068}
  34. }
  35. ```
  36. ---
  37. ## 🧪 Reproduction
  38. Set up environment variables:
  39. ```
  40. echo '
  41. export FIRECRAWL_API_KEY="your_firecrawl_api_key"
  42. export GOOGLE_API_KEY="your_google_api_key"
  43. export HF_HOME="your_hf_home_path"
  44. export OPENROUTER_API_KEY="your_openrouter_api_key"
  45. export SEARCH_ENGINE_ID="your_search_engine_id"
  46. export CHUNKR_API_KEY="your_chunkr_api_key"
  47. ' >> ~/.bashrc && source ~/.bashrc
  48. ```
  49. Create environment:
  50. ```
  51. cd Anemoi
  52. /usr/bin/python3.12 -m venv venv
  53. source venv/bin/activate
  54. pip install -r requirements.txt
  55. ```
  56. We made some minor modifications to CAMEL 0.2.70 for our experiments:
  57. ```
  58. rm -rf venv/lib/python3.12/site-packages/camel
  59. cp -r utils/camel venv/lib/python3.12/site-packages/
  60. ```
  61. Run the experiment:
  62. ```
  63. cd ..
  64. ./gradlew run --console=plain
  65. ```

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.