ROMA: Recursive Open Meta-Agents

Building hierarchical high-performance multi-agent systems made easy! (Beta)

Technical Blog • Paper (Coming soon) • Build Agents for $$$

---

📖 Documentation

• 🚀 Introduction - Understand the vision and architecture behind ROMA

• 📦 Setup - Detailed configuration options and environment setup

• 🤖 Agents Guide - Learn how to create and customize your own agents

• ⚙️ Configuration - Detailed configuration options and environment setup

• 🗺️ Roadmap - See what's coming next for ROMA

🎯 What is ROMA?

ROMA is a meta-agent framework that uses recursive hierarchical structures to solve complex problems. By breaking down tasks into parallelizable components, ROMA enables agents to tackle sophisticated reasoning challenges while maintaining transparency that makes context-engineering and iteration straightforward. The framework offers parallel problem solving where agents work simultaneously on different parts of complex tasks, transparent development with a clear structure for easy debugging, and proven performance demonstrated through our search agent's strong benchmark results. We've shown the framework's effectiveness, but this is just the beginning. As an open-source and extensible platform, ROMA is designed for community-driven development, allowing you to build and customize agents for your specific needs while benefiting from the collective improvements of the community.

🏗️ How It Works

ROMA framework processes tasks through a recursive plan–execute loop:

def solve(task):
    if is_atomic(task):                 # Step 1: Atomizer
        return execute(task)            # Step 2: Executor
    else:
        subtasks = plan(task)           # Step 2: Planner
        results = []
        for subtask in subtasks:
            results.append(solve(subtask))  # Recursive call
        return aggregate(results)       # Step 3: Aggregator

# Entry point:
answer = solve(initial_request)

1. Atomizer – Decides whether a request is atomic (directly executable) or requires planning.
2. Planner – If planning is needed, the task is broken into smaller subtasks. Each subtask is fed back into the Atomizer, making the process recursive.
3. Executors – Handle atomic tasks. Executors can be LLMs, APIs, or even other agents — as long as they implement an agent.execute() interface.
4. Aggregator – Collects and integrates results from subtasks. Importantly, the Aggregator produces the answer to the original parent task, not just raw child outputs.

📐 Information Flow

• Top-down: Tasks are decomposed into subtasks recursively.
• Bottom-up: Subtask results are aggregated upwards into solutions for parent tasks.
• Left-to-right: If a subtask depends on the output of a previous one, it waits until that subtask completes before execution.

This structure makes the system flexible, recursive, and dependency-aware — capable of decomposing complex problems into smaller steps while ensuring results are integrated coherently.

Click to view the system flow diagram

flowchart TB
    A[Your Request] --> B{Atomizer}
    B -->|Plan Needed| C[Planner]
    B -->|Atomic Task| D[Executor]

    %% Planner spawns subtasks
    C --> E[Subtasks]
    E --> G[Aggregator]

    %% Recursion
    E -.-> B  

    %% Execution + Aggregation
    D --> F[Final Result]
    G --> F

    style A fill:#e1f5fe
    style F fill:#c8e6c9
    style B fill:#fff3e0
    style C fill:#ffe0b2
    style D fill:#d1c4e9
    style G fill:#c5cae9

Loading

🚀 30-Second Quick Start

git clone https://github.com/sentient-agi/ROMA.git
cd ROMA

# Run the automated setup
./setup.sh

Choose between:

• Docker Setup (Recommended) - One-command setup with isolation
• Native Setup - Direct installation for development

🛠️ Technical Stack

• Framework: Built on AgnoAgents
• Backend: Python 3.12+ with FastAPI/Flask
• Frontend: React + TypeScript with real-time WebSocket
• LLM Support: Any provider via LiteLLM
• Data Persistence: Enterprise S3 mounting with security validation
  • 🔒 goofys FUSE mounting for zero-latency file access
  • 🛡️ Path injection protection with comprehensive validation
  • 🔐 AWS credentials verification before operations
  • 📁 Dynamic Docker Compose with secure volume mounting
• Code Execution: E2B sandboxes with unified S3 integration
• Security: Production-grade validation and error handling
• Features: Multi-modal, tools, MCP, hooks, caching

📦 Installation Options

Quick Start (Recommended)

# Main setup (choose Docker or Native)
./setup.sh

# Optional: Setup E2B sandbox integration
./setup.sh --e2b

# Test E2B integration  
./setup.sh --test-e2b

Command Line Options

./setup.sh --docker     # Run Docker setup directly
./setup.sh --docker-from-scratch  # Rebuild Docker images/containers from scratch (down -v, no cache)
./setup.sh --native     # Run native setup directly (macOS/Ubuntu/Debian)
./setup.sh --e2b        # Setup E2B template (requires E2B_API_KEY + AWS creds)
./setup.sh --test-e2b   # Test E2B template integration
./setup.sh --help       # Show all available options

Manual Installation

See setup docs for detailed instructions.

🏗️ Optional: E2B Sandbox Integration

For secure code execution capabilities, optionally set up E2B sandboxes:

# After main setup, configure E2B (requires E2B_API_KEY and AWS credentials in .env)
./setup.sh --e2b

# Test E2B integration
./setup.sh --test-e2b

E2B Features:

• 🔒 Secure Code Execution - Run untrusted code in isolated sandboxes
• ☁️ S3 Integration - Automatic data sync between local and sandbox environments
• 🚀 goofys Mounting - High-performance S3 filesystem mounting
• 🔧 AWS Credentials - Passed securely via Docker build arguments

🤖 Pre-built Agents

Note: These agents are demonstrations built using ROMA's framework through simple vibe-prompting and minimal manual tuning. They showcase how easily you can create high-performance agents with ROMA, rather than being production-final solutions. Our mission is to empower the community to build, share, and get rewarded for creating innovative agent recipes and use-cases.

ROMA comes with example agents that demonstrate the framework's capabilities:

🔍 General Task Solver

A versatile agent powered by ChatGPT Search Preview for handling diverse tasks:

• Intelligent Search: Leverages OpenAI's latest search capabilities for real-time information
• Flexible Planning: Adapts task decomposition based on query complexity
• Multi-Domain: Handles everything from technical questions to creative projects
• Quick Prototyping: Perfect for testing ROMA's capabilities without domain-specific setup

Perfect for: General research, fact-checking, exploratory analysis, quick information gathering

🔬 Deep Research Agent

A comprehensive research system that breaks down complex research questions into manageable sub-tasks:

• Smart Task Decomposition: Automatically splits research topics into search, analysis, and synthesis phases
• Parallel Information Gathering: Executes multiple searches simultaneously for faster results
• Multi-Source Integration: Combines results from web search, Wikipedia, and specialized APIs
• Intelligent Synthesis: Aggregates findings into coherent, well-structured reports

Perfect for: Academic research, market analysis, competitive intelligence, technical documentation

💹 Crypto Analytics Agent

Specialized financial analysis agent with deep blockchain and DeFi expertise:

• Real-Time Market Data: Integrates with Binance, CoinGecko, and DefiLlama APIs
• On-Chain Analytics: Access to Arkham Intelligence for wallet tracking and token flows
• Technical Analysis: Advanced charting with OHLC data and market indicators
• DeFi Metrics: TVL tracking, yield analysis, protocol comparisons
• Secure Execution: Runs analysis in E2B sandboxes with data persistence

Perfect for: Token research, portfolio analysis, DeFi protocol evaluation, market trend analysis

All three agents demonstrate ROMA's recursive architecture in action, showing how complex queries that would overwhelm single-pass systems can be elegantly decomposed and solved. They serve as templates and inspiration for building your own specialized agents.

Your First Agent in 5 Minutes

./setup.sh  # Automated setup with Docker or native installation

Access all the pre-defined agents through the frontend on localhost:3000 after setting up the backend on localhost:5000. Please checkout Setup and the Agents guide to get started!

# Your first agent in 3 lines
from sentientresearchagent import SentientAgent

agent = SentientAgent.create()
result = await agent.run("Create a podcast about AI safety")

📊 Benchmarks

We evaluate our simple implementation of a search system using ROMA, called ROMA-Search across three benchmarks: SEAL-0, FRAMES, and SimpleQA.
Below are the performance graphs for each benchmark.

SEAL-0

SealQA is a new challenging benchmark for evaluating Search-Augmented Language models on fact-seeking questions where web search yields conflicting, noisy, or unhelpful results.

---

FRAMES

View full results

A comprehensive evaluation dataset designed to test the capabilities of Retrieval-Augmented Generation (RAG) systems across factuality, retrieval accuracy, and reasoning.

---

SimpleQA

View full results

Factuality benchmark that measures the ability for language models to answer short, fact-seeking questions.

✨ Features

🔄 Recursive Task Decomposition Automatically breaks down complex tasks into manageable subtasks with intelligent dependency management. Runs independent sub-tasks in parallel.	🤖 Agent Agnostic Works with any provider (OpenAI, Anthropic, Google, local models) through unified interface, as long as it has an agent.run() command, then you can use it!
🔍 Complete Transparency Stage tracing shows exactly what happens at each step - debug and optimize with full visibility	🔌 Connect Any Tool Seamlessly integrate external tools and protocols with configurable intervention points. Already includes production-grade connectors such as E2B, file-read-write, and more.

🙏 Acknowledgments

This framework would not have been possible if it wasn't for these amazing open-source contributions!

• Inspired by the hierarchical planning approach described in "Beyond Outlining: Heterogeneous Recursive Planning" by Xiong et al.
• Pydantic - Data validation using Python type annotations
• Agno - Framework for building AI agents
• E2B - Cloud runtime for AI agents

📚 Citation

If you use the ROMA repo in your research, please cite:

@software{al_zubi_2025_17052592,
  author       = {Al-Zubi, Salah and
                  Nama, Baran and
                  Kaz, Arda and
                  Oh, Sewoong},
  title        = {SentientResearchAgent: A Hierarchical AI Agent
                   Framework for Research and Analysis
                  },
  month        = sep,
  year         = 2025,
  publisher    = {Zenodo},
  version      = {ROMA},
  doi          = {10.5281/zenodo.17052592},
  url          = {https://doi.org/10.5281/zenodo.17052592},
  swhid        = {swh:1:dir:69cd1552103e0333dd0c39fc4f53cb03196017ce
                   ;origin=https://doi.org/10.5281/zenodo.17052591;vi
                   sit=swh:1:snp:f50bf99634f9876adb80c027361aec9dff97
                   3433;anchor=swh:1:rel:afa7caa843ce1279f5b4b29b5d3d
                   5e3fe85edc95;path=salzubi401-ROMA-b31c382
                  },
}

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.