ADD1xD/ClearShield
GitHub: ADD1xD/ClearShield
Stars: 2 | Forks: 0
# ⚔️ ClearShield
## 📖 Introduction & Vision
**ClearShield** is a premium, real-time threat intelligence network for EVM-compatible blockchains.
Unlike traditional centralized security registries that can be censored, targeted, or manipulated, ClearShield stores all malicious address registries on **Arkiv**—a highly optimized, decentralized data layer. Once a report is submitted, no malicious actor can delete their digital footprint, and no centralized platform can censor a warning.
Every report and audit confirmation is bound cryptographically to its respective reporter using Braga blockchain transaction signatures, utilizing immutable on-chain attribution.
## 🛡️ Core Features
1. **AI-Driven Automated Verification & Scoring**
* Before being recorded, community reports are validated in real-time by a multi-model **Gemini AI service**.
* The AI filters out test entries/spam, derives a **severity score (1–10)**, and writes a concise **2-sentence plain-English summary** of the exploit vectors to save user assets.
2. **Digital Samurai Guardian Aesthetic System**
* A premium, high-energy layout styling featuring a dark, tactical color palette (crimson intensity, matte stealth blacks, and glowing cybersecurity teals).
* Visual enhancements include: a fine film-grain overlay, moving laser grids, horizontal and vertical katana-slash accents, glowing corner brackets, and giant, ambient background watermarks representing the virtues of **Bushido** (`義` Integrity, `勇` Courage, `礼` Respect, `忠` Loyalty).
3. **Decentralized Community Audit & Confirmation Feed**
* Live streaming of global threat logs.
* Connect MetaMask to Braga testnet to "+ Confirm" reports you have also personally witnessed. The audit count increases dynamically in real-time to alert others.
4. **Production-Ready Developer & Agent APIs**
* A high-performance, rate-limited `/api/v1/check` endpoint designed for hot transaction paths in dApps, wallets, and autonomous AI agents.
## 🏛️ Decentralized Data Architecture
ClearShield stores its intelligence using two primary entity schemas published to the Braga testnet under the global namespace `CLEARSHIELD_V1`.
### 1. Entity Types
| Entity Type | Description | TTL (Decay Window) |
| :--- | :--- | :--- |
| `threat_report` | Original incident record detailing the malicious address, description, chain, and estimated USD loss. | **90 Days** (Unconfirmed reports decay automatically) |
| `threat_confirmation` | An audit signature from another wallet validating the existence of the threat. | **90 Days** (Co-expires with the parent report) |
### 🔑 Cryptographic Attribution & Security Metadata
Every block record on the Arkiv data layer is natively sealed with two key metadata properties that control authorization:
* **`creator` (`$creator` - Immutable)**: The wallet address that originally signed the transaction. This provides **tamper-proof attribution** for threat reporting and confirmations. In the frontend, this guarantees that confirmations cannot be sybil-spammed by a single wallet.
* **`owner` (`$owner` - Mutable)**: The administrative controller address of the entity, allowing the report author to update or delete the record if the threat profile shifts.
ClearShield retrieves these properties directly from the root-level properties of the returned entity objects using the `@arkiv-network/sdk` and maps them cleanly into the UI:
const creator = entity.creator; // Fully utilized for security attribution!
## 🛠️ Tech Stack & Dependencies
* **Frontend & Backend APIs:** Next.js 16.2 (App Router) + TypeScript + React 19
* **Styling & Theme Engine:** Vanilla CSS + custom design tokens
* **Decentralized Data Layer:** `@arkiv-network/sdk` (indexing Braga testnet events via `subscribeEntityEvents`)
* **Core Ledger Network:** Arkiv Braga Testnet
* **Generative AI Core:** `@google/generative-ai` (Gemini Flash & Fallbacks for validation and synthesis)
* **Web3 Integration:** Wagmi + Viem + MetaMask custom providers
## 🚀 Setup & Installation
### Prerequisites
* **Node.js:** version `18.0.0` or higher
* **MetaMask** installed in your web browser
* **Braga GLM Test Tokens** (Available at the [Arkiv Braga Faucet](https://braga.hoodi.arkiv.network/faucet/))
### 1. Clone & Install
git clone https://github.com/your-username/clearshield.git
cd clearshield
npm install
### 2. Environment Configuration
Copy the template variables file:
cp .env.local.example .env.local
Fill in the following credentials inside `.env.local`:
# Gemini API credentials (from https://aistudio.google.com/)
GEMINI_API_KEY=AIzaSy...
# Braga Node RPC and Websocket Configurations
NEXT_PUBLIC_ARKIV_RPC=https://braga.hoodi.arkiv.network/rpc
NEXT_PUBLIC_ARKIV_WS=wss://braga.hoodi.arkiv.network/rpc/ws
NEXT_PUBLIC_ARKIV_CHAIN_ID=60138453102
# Server-side wallets fallback (Optional)
ARKIV_PRIVATE_KEY=your_private_key_here
### 3. Run Locally
Start the development server:
npm run dev
Open **[http://localhost:3000](http://localhost:3000)** in your browser.
### 4. Braga Testnet Wallet Setup
When clicking **Connect Wallet** inside the web app, MetaMask will automatically prompt you to add and switch to the Arkiv Braga Testnet. If you need to add it manually:
* **Network Name:** `Arkiv Braga Testnet`
* **RPC URL:** `https://braga.hoodi.arkiv.network/rpc`
* **Chain ID:** `60138453102` (Hex: `0xe0087f86e`)
* **Currency Symbol:** `GLM`
* **Block Explorer:** [https://explorer.braga.hoodi.arkiv.network](https://explorer.braga.hoodi.arkiv.network)
## ⚡ Integration API (For Wallets & Agents)
Integrate decentralized threat checks directly into hot transaction paths:
### 1. GET Request
curl -H "x-api-key: YOUR_KEY" "https://clear-shield.vercel.app/api/v1/check?address=0x379a2970755003256c19a4de88b9c05b21ab401f&includeReports=true"
### 2. Response JSON Schema
{
"address": "0x379a2970755003256c19a4de88b9c05b21ab401f",
"isFlagged": true,
"riskLevel": "high",
"confidence": "high",
"reportCount": 2,
"confirmationCount": 9,
"maxSeverity": 7,
"dominantThreatType": "approval_drain",
"totalAmountLostUSD": 0,
"verdict": "This address has been flagged for active approval drain interactions. Nine distinct wallets have cryptographically co-signed this alert. Avoid interacting or signing any deposit methods with this contract.",
"reports": [
{
"entityKey": "0x67a165a859f710683f56549bd9d62b8e60371216b745a8a4951a27e477df7b91",
"threatType": "approval_drain",
"severityScore": 7,
"confirmationCount": 4,
"amountLostUSD": 0,
"reportedAt": 1779548385904,
"aiSummary": "Contract initiates unlimited approval requests for ERC-20 assets under the guise of an aggregator utility."
}
]
}
Detailed integration templates for JavaScript, Python, Ethers.js and Autonomous Agent system prompts are fully documented in [docs/INTEGRATION.md](docs/INTEGRATION.md).
## ⚖️ License
This project is licensed under the MIT License — see the [LICENSE](LICENSE) file for details.
标签:自动化攻击