About StealthGuard 402
StealthGuard 402 is an open-source MCP server for agents that need web extraction, market signals, and pre-trade checks. Its tools combine anti-bot scraping with prompt-injection filtering, Polymarket order-flow analysis, and contract and liquidity checks.
Key features
- 01Anti-bot web scraping and content extraction
- 02Prompt-injection filtering of scraped content
- 03Polymarket odds and whale-flow analysis
- 04Base and Solana pre-trade security checks
Use cases
- 01Collect web content while filtering injection traps
- 02Compare prediction odds with spot momentum and order flow
- 03Check token contracts and DEX liquidity before trading
MCP support
The server exposes three tools over stdio and Streamable HTTP: `stealth_scrape_and_sanitize`, `polymarket_alpha_divergence`, and `pre_trade_agent_gate`. A hosted remote MCP endpoint is also documented.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://stealthguard-402.onrender.com/mcpclaude mcp add --transport http ak10082247-max-stealthguard-402 https://stealthguard-402.onrender.com/mcp{"mcpServers":{"ak10082247-max-stealthguard-402":{"type":"http","url":"https://stealthguard-402.onrender.com/mcp"}}}Open in Cursor{"servers":{"ak10082247-max-stealthguard-402":{"type":"http","url":"https://stealthguard-402.onrender.com/mcp"}}}Open in VS Code{"mcpServers":{"ak10082247-max-stealthguard-402":{"type":"http","url":"https://stealthguard-402.onrender.com/mcp"}}}{"mcpServers":{"ak10082247-max-stealthguard-402":{"serverUrl":"https://stealthguard-402.onrender.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
15| Parameter | Type | Required | Description |
|---|---|---|---|
| url | string | Yes | Target destination URL |
| render_js | boolean | No | Enable stealth headless browser rendering |
| strict_mode | boolean | No | Deterministic sanitization strictness |
Details
From the project READMEStealthGuard 402 🛡️ — Open-Source MCP Server for Web Scraping & Market Intelligence
This repository contains the complete open-source Model Context Protocol (MCP) server implementation of StealthGuard 402. It equips LLM agents (Claude Desktop, Cursor, Windsurf, LangChain) with tools for Cloudflare-bypassing web scraping, prompt injection firewall sanitization, Polymarket whale orderbook divergence, and cross-chain pre-trade agent gating.
🛠️ MCP Tools Implemented
This MCP server provides the following tools via standard stdio and streamable HTTP transports:
| Tool Name | Description | Key Parameters |
|---|---|---|
stealth_scrape_and_sanitize | Bypasses Cloudflare Turnstile, Akamai, and anti-bot protection using Scrapling, extracts clean web content, and strips prompt injection traps before LLM context ingestion. | url (string, required), render_js (boolean, default: false), strict_mode (boolean, default: true) |
polymarket_alpha_divergence | Scans live Polymarket prediction markets, calculates divergence between prediction odds and on-chain spot momentum, and tracks whale orderbook flow. | category (string, e.g. crypto, politics, ai), limit (int, default: 25) |
pre_trade_agent_gate | Unified pre-trade security gate for autonomous agents: audits contract bytecode, verifies DEX LP depth on Base and Solana, and sanitizes project metadata. | token_or_service_address (string, required), chain (base or solana, default: base), strict_mode (boolean) |
🚀 Running from Source (Local Stdio Transport)
You can run this MCP server locally from source using Python and the Model Context Protocol SDK.
1. Installation
# Clone the open-source repository
git clone https://github.com/ak10082247-max/stealthguard-402.git
cd stealthguard-402
# Set up virtual environment
python -m venv .venv
# On Linux/macOS:
source .venv/bin/activate
# On Windows:
.\.venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
2. Configure with Claude Desktop
Add this server to your claude_desktop_config.json:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"stealthguard-402": {
"command": "python",
"args": [
"-m",
"src.mcp_server"
],
"cwd": "/absolute/path/to/stealthguard-402"
}
}
}
Or run via uv:
{
"mcpServers": {
"stealthguard-402": {
"command": "uv",
"args": [
"run",
"--directory",
"/absolute/path/to/stealthguard-402",
"python",
"-m",
"src.mcp_server"
]
}
}
}
3. Configure with Cursor or Windsurf
Add to your workspace .cursor/mcp.json or Windsurf settings:
{
"mcpServers": {
"stealthguard-402": {
"command": "python",
"args": ["-m", "src.mcp_server"]
}
}
}
🌐 Connecting to the Hosted Remote MCP Endpoint
For cloud deployments and remote agent integrations, StealthGuard is deployed 24/7 on high-availability infrastructure:
- Remote MCP URL:
https://stealthguard-402.onrender.com/mcp - Protocol: Model Context Protocol (Streamable HTTP / SSE JSON-RPC 2.0)
- Discovery Manifest:
https://stealthguard-402.onrender.com/.well-known/mcp/server-card.json - x402 Manifest:
https://stealthguard-402.onrender.com/.well-known/x402.json
🏛️ System Architecture
sequenceDiagram
autonumber
actor LLM as Claude / Agent / Cursor
participant MCP as StealthGuard MCP Server
participant Scrapling as Scrapling Anti-Bot Engine
participant Firewall as CanaryGuard Heuristic Filter
participant Target as Destination Web Page
LLM->>MCP: tools/call ("stealth_scrape_and_sanitize", { url })
MCP->>Scrapling: Fetch Destination (Bypass Cloudflare)
Scrapling->>Target: HTTP Request with Dynamic Fingerprinting
Target-->>Scrapling: HTML Payload
Scrapling->>Firewall: Extract Text & Scan for Injections
Firewall-->>MCP: Sanitized Text + Safety Receipt
MCP-->>LLM: Safe Content Payload (Zero Prompt Leaks)
📦 Developer Integrations & Autonomous Auto-Pay
StealthGuard supports 100% gasless automated settlement on Base Mainnet (eip155:8453) and Solana Mainnet (solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp). Zero native gas ($0 ETH / $0 SOL) is required from your agent wallet.
1. Autonomous Agent Auto-Pay Quickstart (TypeScript)
Wrap standard fetch with @x402/fetch to automatically detect HTTP 402, sign EIP-712 payment authorization, and unlock full unredacted scrape & intelligence feeds:
npm install @x402/fetch @x402/core @x402/evm viem
import { wrapFetchWithPayment } from "@x402/fetch";
import { x402Client } from "@x402/core/client";
import { ExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";
const account = privateKeyToAccount(process.env.AGENT_PRIVATE_KEY as `0x${string}`);
const client = new x402Client();
client.register("eip155:8453", new ExactEvmScheme(account));
const xFetch = wrapFetchWithPayment(fetch, client);
const res = await xFetch("https://stealthguard-402.onrender.com/v1/stealth_scrape_and_sanitize", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ url: "https://example.com", strict_mode: true }),
});
const data = await res.json();
console.log("Full Sanitized Content:", data);
2. Autonomous Agent Auto-Pay Quickstart (Python)
AI agent frameworks (LangChain, CrewAI, AutoGen) can auto-settle the 402 challenge in Python:
pip install requests eth-account
import base64, json, secrets, time, requests
from eth_account import Account
from eth_account.messages import encode_typed_data
def paid_stealth_scrape(target_url: str, private_key: str):
account = Account.from_key(private_key)
endpoint = "https://stealthguard-402.onrender.com/v1/stealth_scrape_and_sanitize"
payload = {"url": target_url, "strict_mode": True}
# Step 1: Initial probe (triggers 402 with structured 50% teaser + challenge)
res = requests.post(endpoint, json=payload)
if res.status_code == 200:
return res.json()
# Step 2: Auto-sign EIP-712 transfer authorization (100% gasless on Base)
raw_pr = res.headers.get("PAYMENT-REQUIRED") or res.headers.get("X-Payment-Required")
pr_json = json.loads(base64.b64decode(raw_pr).decode("utf-8"))
accepted = next(a for a in pr_json["accepts"] if "8453" in str(a.get("network", "")))
now = int(time.time())
nonce = secrets.token_bytes(32)
types = {"TransferWithAuthorization": [
{"name": "from", "type": "address"}, {"name": "to", "type": "address"},
{"name": "value", "type": "uint256"}, {"name": "validAfter", "type": "uint256"},
{"name": "validBefore", "type": "uint256"}, {"name": "nonce", "type": "bytes32"},
]}
domain = {"name": "USD Coin", "version": "2", "chainId": 8453, "verifyingContract": accepted["asset"]}
msg = {"from": account.address, "to": accepted["payTo"], "value": int(accepted["amount"]), "validAfter": 0, "validBefore": now + 300, "nonce": nonce}
sig = account.sign_message(encode_typed_data(domain_data=domain, message_types=types, message_data=msg)).signature.hex()
payment_b64 = base64.b64encode(json.dumps({
"x402Version": 2, "resource": pr_json.get("resource"), "accepted": accepted,
"payload": {"signature": "0x" + sig.lstrip("0x"), "authorization": {
"from": account.address, "to": accepted["payTo"], "value": str(accepted["amount"]),
"validAfter": "0", "validBefore": str(now + 300), "nonce": "0x" + nonce.hex()
}}
}).encode()).decode()
# Step 3: Replay with PAYMENT-SIGNATURE
paid_res = requests.post(endpoint, json=payload, headers={"PAYMENT-SIGNATURE": payment_b64})
return paid_res.json()
3. Direct REST API (cURL)
# Returns 402 with structured 50% free preview + unlock instructions:
curl -X POST https://stealthguard-402.onrender.com/v1/stealth_scrape_and_sanitize \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com", "strict_mode": true}'
# Polymarket whale orderflow
curl -X POST https://stealthguard-402.onrender.com/v1/polymarket-odds-pulse \
-H "Content-Type: application/json" \
-d '{"category": "crypto"}'
💳 Optional Cloud Subscriptions (MCPize / Apify)
For teams seeking turnkey managed infrastructure with flat monthly credit card billing rather than self-hosting:
- MCPize Directory: Connect via mcpize.com/dashboard
- Apify Pay-Per-Event: Actor distribution on apify.com
📄 License
This open-source project is licensed under the MIT License — see the LICENSE file for details.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.