⚡ Polygon x402 Autonomous AI Agent Suite & Spend Firewall (v2.5.5)
"High-Throughput Clean Web Markdown, YouTube Intelligence & Cryptographic Web3 Oracles for Autonomous AI Agents & Swarms."
Zero credit cards required. Pure B2A (Business-to-Agent) USDC micropayments & pre-funded vaults on Polygon, Base, and Arbitrum with on-chain EIP-712 spend policy attestations.
🌐 Live Production Gateway & Endpoints
- Production Gateway:
https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app - Agent LLM Standard (
llms.txt):https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/llms.txt - A2A Agent Manifest:
/.well-known/agent.json - OpenAI Plugin Manifest:
/.well-known/ai-plugin.json - MCP Server Card:
/.well-known/mcp/server-card.json - Arbitrage ROI Calculator:
/api/v1/agent/arbitrage-roi - Interactive Studio Dashboard:
https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/dashboard - Interactive OpenAPI Specs:
https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/docs
⚡ 10-Second Instant Test (Zero Auth / 3 Free Calls Included)
No wallet, credit card, or API key needed. Attach an X-Agent-Nonce header to test immediately:
# Instant JSON Request via curl
curl -X POST "https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/api/v1/clean-web" \
-H "Content-Type: application/json" \
-H "X-Agent-Nonce: instant-trial-uuid-001" \
-d '{"url": "https://news.ycombinator.com"}'
# Zero-dependency Python snippet
import urllib.request, json
req = urllib.request.Request(
"https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/api/v1/clean-web",
data=json.dumps({"url": "https://news.ycombinator.com"}).encode(),
headers={"Content-Type": "application/json", "X-Agent-Nonce": "instant-trial-01"}
)
res = json.loads(urllib.request.urlopen(req).read())
print(f"Tokens Saved: {res.get('estimated_token_savings', '87%')}")
print("Clean Markdown Preview:\n", res["clean_markdown"][:300])
💡 Mathematical Token Arbitrage Proof (Economic Rationality)
Autonomous agents route web traffic through x402 because it mathematically saves the operator money on every query:
- Direct Scraping Cost: 80KB HTML ≈ 20,000 tokens ($0.05 USD input cost on GPT-4o / Claude 3.5 Sonnet).
- x402 CleanWeb Cost: $0.001 API fee + 2,600 clean tokens ($0.0065 USD) = $0.0075 USD Total.
- Net Savings: $0.0425 USD saved per web query (85% net dollar reduction).
- Service Fee ROI: 4,250% ROI on the $0.001 fee.
- Live Arbitrage Endpoint:
GET /api/v1/agent/arbitrage-roi?input_tokens=50000
{
"status": "economically_optimal",
"routing_recommendation": "ROUTE_VIA_X402",
"financial_analysis": {
"cost_raw_direct_usd": 0.125,
"cost_with_x402_usd": 0.01725,
"net_dollar_savings": 0.10775,
"net_savings_percentage": "86.2%",
"service_fee_roi_percentage": "10775.0%"
}
}
🔮 14 Production Agent Tools & Micro-Pricing (USDC)
| Tool Name | Endpoint / Function | Cost (USDC) | What It Does |
|---|---|---|---|
| 🌐 Clean Web | GET/POST /api/v1/clean-web |
0.001 USDC | Strips HTML boilerplate, ads, scripts; returns ad-free Markdown (87% token reduction) |
| 📝 Clean Text | GET /api/v1/clean-text |
0.001 USDC | Ultra-lightweight raw plain text for vector/RAG embeddings |
| 🗺️ Site Mapper | GET /api/v1/map-site |
0.002 USDC | Recursive sitemap and internal URL tree discovery |
| 🔍 Agent Search | GET /api/v1/search |
0.002 USDC | Fast real-time keyword search & verified snippets (Tavily engine) |
| 📑 PDF Extractor | GET /api/v1/clean-pdf |
0.005 USDC | Formula- and table-preserved academic paper & research PDF extractor |
| 📦 Batch Clean | POST /api/v1/batch-clean |
0.005 USDC | Concurrent parallel scraping for up to 10 URLs in a single request |
| 🎬 YouTube AI | GET /api/v1/clean-youtube |
0.010 USDC | Gemini 3.6 Flash multimodal video intelligence & timestamped audio transcript |
| 📊 Extract JSON | POST /api/v1/extract-json |
0.030 USDC | Converts arbitrary web content into strict schema-constrained JSON |
| 🔮 Web3 Oracle | POST /api/v1/oracle/grounding |
0.035 USDC | Live Search + Clean-to-JSON + EIP-712 Cryptographic On-Chain Attestation |
| 🛡️ Oracle Verify | POST /api/v1/oracle/verify |
0.000 USDC | Verifies ECDSA signature of CleanWeb Oracle attestations on-chain |
| 🧠 Deep Research | GET /api/v1/deep-research |
0.150 USDC | Multi-source synthesized AI executive research briefing |
| 💼 Vault Deposit | POST /api/v1/vault/deposit |
0.000 USDC | Multi-chain USDC vault prefunding for sub-5ms gasless executions |
| 💳 Vault Balance | GET /api/v1/vault/balance |
0.000 USDC | Real-time query of credit pass balance and validity |
| 🎫 Pass Status | GET /api/v1/pass-status |
0.000 USDC | Inspects status and remaining credits of a prepaid pass |
💻 1-Click MCP Setup (Claude Desktop, Cursor, Windsurf)
Connect CleanWeb directly to Claude Desktop, Cursor, or any Model Context Protocol client using uvx:
{
"mcpServers": {
"polygon-x402-cleanweb": {
"command": "uvx",
"args": ["x402-cleanweb-agent"]
}
}
}
Or configure via local clone:
{
"mcpServers": {
"polygon-x402-cleanweb": {
"command": "python",
"args": ["-u", "mcp_server.py"],
"env": {
"POLYGON_RPC_URL": "https://polygon-bor-rpc.publicnode.com",
"SERVER_WALLET_ADDRESS": "0x255F9991233f86B29dB847c8d5b8CB9915e80dCf",
"USDC_CONTRACT_ADDRESS": "0x3c499c542cEF5E3811e1192ce70d8cC03d5c3359"
}
}
}
}
🤖 Multi-Agent Framework Integrations
Drop-in ready integrations for every major autonomous agent framework via GET /api/v1/agent/integrations/{framework}:
🦜 LangChain / LangGraph
from langchain.tools import tool
import requests
@tool
def clean_web(url: str) -> str:
"""Fetches a URL and returns ad-free, token-optimized Markdown (87% token savings)."""
res = requests.get(
"https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/api/v1/clean-web",
params={"url": url},
headers={"X-Agent-Nonce": "langchain-agent-session"}
)
return res.json().get("clean_markdown", "")
👥 CrewAI
from crewai.tools import tool
import requests
@tool("CleanWeb Tool")
def clean_web(url: str) -> str:
"""Strips HTML boilerplate and returns pure Markdown to save LLM context window."""
res = requests.post(
"https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/api/v1/clean-web",
json={"url": url},
headers={"X-Agent-Nonce": "crewai-agent-session"}
)
return res.json().get("clean_markdown", "")
🤖 Microsoft AutoGen
import requests
def clean_web_tool(url: str) -> str:
res = requests.get(
"https://x402-cleanweb-agent-7qxtp3324q-du.a.run.app/api/v1/clean-web",
params={"url": url},
headers={"X-Agent-Nonce": "autogen-session"}
)
return res.json().get("clean_markdown", "")
# assistant.register_for_llm(name="clean_web", description="Clean web markdown")(clean_web_tool)
🏛️ Verified On-Chain Smart Contracts (Polygon, Base, Arbitrum)
CleanWeb operates verified, deterministic smart contracts for on-chain EIP-712 oracle verification and autonomous agent pre-funded vault management.
| Contract Name | Polygon Mainnet (137) ✔ | Base Mainnet (8453) ✔ | Arbitrum One (42161) ✔ |
|---|---|---|---|
AgentPaymentVault |
0x45ecBf...1861 |
0x28292D...76DD |
0x28292D...76DD |
CleanWebOracleConsumer |
0xAECbfB...6D66 |
0x2394d8...dabe |
0x2394d8...dabe |
CleanWebOracleVerifier |
0x18fA45...Dc46 |
0x3eD259...0740 |
0x3eD259...0740 |
🧪 Production Readiness Review: 43/43 Tests Passed (100%)
Verified across all 38+ FastAPI routes, 14 agent tools, multi-chain settlement vaults, and EIP-712 attestations:
python -m pytest tests/ -v
tests/test_agent_discovery.py::test_well_known_manifests PASSED [ 2%]
tests/test_agent_discovery.py::test_agent_capabilities_reflection PASSED [ 4%]
tests/test_agent_discovery.py::test_agent_pricing_catalog PASSED [ 6%]
tests/test_agent_discovery.py::test_agent_arbitrage_roi PASSED [ 9%]
tests/test_agent_discovery.py::test_agent_framework_integrations PASSED [ 11%]
tests/test_diagnostics.py::test_run_full_diagnostic PASSED [ 13%]
tests/test_diagnostics.py::test_diagnostics_endpoint PASSED [ 16%]
tests/test_enhanced_agent_features.py (16 tests) PASSED [ 53%]
tests/test_oracle_grounding.py (3 tests) PASSED [ 60%]
tests/test_payment_comprehensive.py (4 tests) PASSED [ 69%]
tests/test_phase1_cleaners.py (4 tests) PASSED [ 79%]
tests/test_phase2_payments.py (5 tests) PASSED [ 90%]
tests/test_routes_coverage.py::test_all_api_routes_operational PASSED [ 93%]
tests/test_treasury.py (3 tests) PASSED [100%]
================== 43 passed, 2 warnings in 85.93s (100%) ==================
⚖️ Legal, Fair Use & Financial Non-Liability Disclaimer
CleanWeb Studio operates strictly under the Transformative Non-Expressive Text/Data Mining (TDM) Fair Use doctrine for autonomous AI reasoning.
- No Financial or Investment Warranty: All oracle feeds, search results, and scraped contents are provided AS-IS. CleanWeb Studio assumes zero liability for DeFi smart contract liquidations, prediction market settlements (e.g. Polymarket), or trading losses resulting from external web misreporting or LLM hallucinations.
- Caller Responsibility: The autonomous agent deployer/caller is solely responsible for respecting target website intellectual property and applicable laws.
- Zero-Data Retention (GDPR Compliant): All fetched HTML and raw assets are processed ephemerally in RAM and purged immediately upon response transmission.
- 100% OFAC Sanctions Filtering: All micropayment addresses are verified against OFAC Specially Designated Nationals (SDN) lists.
- Full Legal Endpoints:
GET /api/v1/legal/terms: Official Machine-to-Machine Terms of ServiceGET /api/v1/legal/disclaimer: Full Legal & Financial Disclaimer
📜 License
Distributed under the MIT License. See LICENSE for more information.
Metadata
Release files for x402-cleanweb-agent 2.5.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| x402_cleanweb_agent-2.5.5.tar.gz | 99.7 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| x402_cleanweb_agent-2.5.5-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 196.4 kB
Release files / x402_cleanweb_agent-2.5.5.tar.gz
| Download URL | x402_cleanweb_agent-2.5.5.tar.gz |
|---|---|
| Size | 99.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f36547af472240d12ad10bfe84b7202d31679db1cbbc6fe213d957280222b1a8
|
|
BLAKE2b-256 checksum How to use checksums |
561b6b433317d4a77e82db0da39a6eb53e1aad8f32ef278901ecef5ec5c2f451
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|
Release files / x402_cleanweb_agent-2.5.5-py3-none-any.whl
| Download URL | x402_cleanweb_agent-2.5.5-py3-none-any.whl |
|---|---|
| Size | 96.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4026d2d9a3bae3fc71e4ba8517314c6a4c530bf928b59edd3dc160e80e540523
|
|
BLAKE2b-256 checksum How to use checksums |
58f684f5b5c56a83e0a70e82cce909a586a7de6e037dbf94f2d130dbfa88182d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|