AI-powered airdrop discovery and qualification platform β multi-agent opportunity scanning, task planning, eligibility checking, and reward estimation powered by Xiaomi MiMo.
Airdrop Radar uses a multi-agent AI system to help Web3 users discover, evaluate, and qualify for cryptocurrency airdrop opportunities across multiple blockchain ecosystems. Each agent specializes in a specific aspect of the airdrop lifecycle, orchestrated through a FastAPI backend with a clean dashboard frontend.
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Airdrop Radar Frontend β
β (Dark-themed SPA Dashboard) β
ββββββββββββββββββββββββββββββ¬βββββββββββββββββββββββββββββββββββββ
β REST API
ββββββββββββββββββββββββββββββΌβββββββββββββββββββββββββββββββββββββ
β FastAPI Backend (main.py) β
β Endpoints: /discover /plan /check /estimate /agents/status β
βββββββ¬βββββββββββββββ¬βββββββββββββββ¬βββββββββββββββ¬βββββββββββββββ
β β β β
βββββββΌββββββ ββββββββΌββββββ βββββββΌβββββββ ββββββΌβββββββββββ
β π Opp. β β π Task β β β
Elig. β β π° Reward β
β Scanner β β Planner β β Checker β β Estimator β
β Agent β β Agent β β Agent β β Agent β
βββββββ¬ββββββ ββββββββ¬ββββββ βββββββ¬βββββββ ββββββ¬βββββββββββ
β β β β
βββββββΌβββββββββββββββΌβββββββββββββββΌβββββββββββββββΌβββββββββββββββ
β Xiaomi MiMo LLM (MiMo-7B-RL) β
β Chain-of-thought reasoning for Web3 analysis β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
- π Opportunity Scanner β Scans 6+ blockchain ecosystems (Ethereum, Solana, Base, Arbitrum, Optimism, Polygon) for airdrop opportunities using LLM analysis of protocol activity, funding, and historical patterns
- π Task Planner β Creates gas-optimized, priority-ranked action plans to maximize airdrop eligibility within budget constraints
- β Eligibility Checker β Analyzes wallet addresses against airdrop criteria, classifying eligibility status with actionable recommendations
- π° Reward Estimator β Estimates potential token rewards and USD value ranges using historical airdrop data, protocol valuations, and wallet activity analysis
- RESTful API with full OpenAPI documentation
- Dark-themed responsive dashboard
- Agent status monitoring
- Real-time multi-agent orchestration
- Demo mode (works without API key using generated data)
- CORS-enabled for frontend integration
| Layer | Technology |
|---|---|
| LLM | Xiaomi MiMo (MiMo-7B-RL) |
| Backend | Python 3.11+, FastAPI, Pydantic v2 |
| HTTP Client | httpx (async) |
| Frontend | Vanilla HTML/CSS/JS (no build step) |
| Deployment | Netlify (frontend), any Python host (backend) |
| License | MIT |
- Python 3.11+
- Xiaomi MiMo API key (optional β works in demo mode without one)
cd backend
# Create virtual environment
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
# Configure environment
cp .env.example .env
# Edit .env with your MIMO_API_KEY
# Run the server
uvicorn app.main:app --reload --port 8000The API will be available at http://localhost:8000 with interactive docs at http://localhost:8000/docs.
cd frontend
# Serve with any static file server
python -m http.server 3000
# Or use npx serve .
# Or deploy to NetlifyOpen http://localhost:3000 in your browser.
| Method | Endpoint | Description |
|---|---|---|
POST |
/discover/airdrops |
Scan for airdrop opportunities |
POST |
/plan/tasks |
Generate prioritized task plan |
POST |
/check/eligibility |
Check wallet eligibility |
POST |
/estimate/rewards |
Estimate potential rewards |
GET |
/agents/status |
Get all agent statuses |
GET |
/health |
Health check |
curl -X POST http://localhost:8000/discover/airdrops \
-H "Content-Type: application/json" \
-d '{"chains": ["Ethereum", "Solana"], "min_confidence": 0.5}'curl -X POST http://localhost:8000/estimate/rewards \
-H "Content-Type: application/json" \
-d '{"wallet_address": "0x742d35Cc6634C0532925a3b844Bc9e7595f2bD18"}'All responses use Pydantic models with consistent schemas. See /docs for full OpenAPI specification with request/response models.
airdrop-radar/
βββ backend/
β βββ app/
β β βββ main.py # FastAPI application
β β βββ agents/
β β β βββ __init__.py
β β β βββ opportunity_scanner.py # π Discovery agent
β β β βββ task_planner.py # π Planning agent
β β β βββ eligibility_checker.py # β
Eligibility agent
β β β βββ reward_estimator.py # π° Estimation agent
β β βββ models/
β β βββ __init__.py
β β βββ schemas.py # Pydantic data models
β βββ requirements.txt
β βββ .env.example
βββ frontend/
β βββ index.html # Dashboard SPA
β βββ css/style.css # Dark theme styles
β βββ js/app.js # Frontend logic
βββ proof/
β βββ README.md
β βββ architecture.md
β βββ token_consumption_estimate.md
βββ README.md
βββ LICENSE
βββ .gitignore
βββ netlify.toml
Each agent follows a consistent pattern:
- Receive a structured request
- Build a domain-specific prompt with context
- Query the Xiaomi MiMo LLM with the prompt
- Parse the structured JSON response
- Validate and return Pydantic models
When no API key is configured, agents fall back to realistic generated data for development and testing.
Each agent uses carefully crafted system prompts that:
- Define the agent's role and expertise
- Specify exact output JSON schema
- Provide domain knowledge and analysis frameworks
- Include comparable historical data for calibration
100t-orbit Β· ai-agents Β· airdrop Β· web3 Β· fastapi Β· llm Β· mimo Β· multi-agent Β· reward-estimation Β· xiaomi-mimo
MIT License β see LICENSE for details.
This tool is for educational and research purposes only. It does not constitute financial advice. Always do your own research (DYOR) before participating in any airdrop or DeFi protocol. Airdrop estimates are speculative and based on historical patterns β actual rewards may vary significantly.