Crypto Risk Assessment Tool
“A real-time, multi-chain crypto risk analysis system that integrates data retrieval and signal-based evaluation, using RAG to deliver strictly facts-only, non-advisory, and interpretable insights.”
Install / Use
npx skills add lraut-hub/Crypto-Risk-Assessment-ToolInstalls into whichever agent you are using.
README
🕵️♂️ Crypto Risk Assessment Tool
A high-precision, modular Retrieval-Augmented Generation (RAG) engine designed for deterministic crypto asset risk verification. This tool converts raw blockchain, market, and regulatory data into structured, source-backed signal reports.
🚀 The Problem
In the volatile crypto landscape, retail traders and compliance teams often struggle with:
- Information Overload: Sifting through fragmented data across explorers and social media.
- Speculative Bias: LLMs often "hallucinate" advice or provide subjective analysis.
- Hidden Risks: Undetected honeypots, predatory taxes, or missing audits in obscure tokens.
The Solution: A fact-only, signal-first verification engine that prioritizes precision and transparency over conversational interaction.
🏗️ Architecture: The 9-Step Pipeline
The system follows a strict, deterministic workflow to ensure every risk report is grounded in verifiable data:
- Query Router: Classifies intent to block advisory/out-of-scope queries.
- Entity Resolver: Maps tickers/names to canonical IDs via CoinGecko.
- Asset Classifier: Identifies asset types (Smart Contract vs. Native Asset).
- Signal Engine: Concurrently fetches data from:
- GoPlus Security: Smart contract vulnerabilities and tax flags.
- CoinGecko: Market liquidity, rank, and exchange listings.
- Regulatory Lists: Daily scraped data from FCA, SEC, and MAS.
- Signal Normalizer: Converts raw numbers into human-readable indicators (e.g.,
⚠️ PREDATORY TAX). - RAG Retriever: Pulls context from ChromaDB for signal grounding.
- LLM Explanation: Uses Llama 3.3 (Groq) to generate a 2-3 line factual summary.
- Output Formatter: Builds a structured Markdown/JSON report.
- Post-Guards: Final regex check to ensure no advisory language is present.
🛡️ Key Signal Layers
| Layer | Indicators Tracked | | :--- | :--- | | Structural | Honeypots, Minting permissions, Proxy/Upgradable contracts, Buy/Sell taxes. | | Market | Market Cap, Trading Volume, Exchange Listings, Market Rank. | | Audit | Security audit status, Verified source code, Known audit providers. | | Regulatory | Warnings from global regulators (FCA, SEC, MAS), Sanctioned status. |
🛠️ Tech Stack
- Backend: FastAPI (Python 3.10+)
- Frontend: React 19 + Vite + TypeScript + TailwindCSS
- AI/LLM: Groq Cloud (
llama-3.3-70b-versatile) - Vector DB: ChromaDB (Embeddings:
BAAI/bge-large-en-v1.5) - Data APIs: GoPlus Security API, CoinGecko API
🏁 Getting Started
Prerequisites
- Python 3.10+
- Node.js 18+
- API Keys: Groq, CoinGecko (optional), GoPlus
Installation
- Clone the Repository:
git clone https://github.com/lraut-hub/Crypto-Risk-Assessment-Tool.git cd Crypto-Risk-Assessment-Tool - Backend Setup:
cd backend pip install -r requirements.txt python main.py - Frontend Setup:
cd ../frontend npm install npm run dev
⚠️ Disclaimer
Facts-only. No investment advice. This tool reports historical and technical signals; it does not predict future performance or guarantee safety. Always perform your own research (DYOR).
Built with precision for the next generation of crypto security.
Related Skills
node-connect
385.5kDiagnose OpenClaw Android, iOS, or macOS node pairing, QR/setup code, route, auth, and connection failures.
blender-python-addon
40.5kBlender Python add-on rules for operators, panels, properties, registration, testing, and API-safe scripting
flutter-development-guidelines-cursorrules-prompt-file
40.5kCursor rules for Flutter development with MVVM architecture, Riverpod state management, Material widgets, and Dart style guidelines.
commit-push-pr
140.6kCommit, push, and open a PR
