Skip to main content
Glama

WebLens MCP

Give Claude, Cursor or any MCP client four tools for reading the web, plus RugProbe to check a Base token before trading. Each call is paid in USDC on Base via x402. There is no account, no API key and no subscription.

Tool

What it returns

Price

weblens_metadata

Title, description, Open Graph, JSON-LD, favicon

$0.002

weblens_links

All links, split into internal and external

$0.002

weblens_extract

Main content as clean Markdown (no ads, menus or popups)

$0.005

weblens_structure

JSON that follows your JSON Schema, from a URL or text

$0.02

rugprobe_check

RugProbe: is this Base token safe to trade? Simulated buy/sell (honeypot, real taxes), liquidity, owner powers, 0-100 risk score

$0.01

rugprobe_report

RugProbe Report: everything above plus the project website (does it list this contract? domain age, socials), a clear/caution/avoid verdict and a short summary

$0.05

rugprobe_position_size

Stop-loss and position size from account risk % and ATR

$0.002

rugprobe_gas

Base gas advisor: submit now or wait, fee suggestions, USD cost

$0.002

weblens_csv_to_json

CSV to typed JSON rows (snake_case keys, column types)

$0.002

weblens_json_repair

Fix broken LLM JSON and validate it against a schema

$0.002

Setup

  1. Create a separate wallet for this and put a few dollars of USDC on Base in it. Never use your main wallet.

  2. Add the server to your MCP client. For Claude Desktop, put this in claude_desktop_config.json:

{
  "mcpServers": {
    "weblens": {
      "command": "npx",
      "args": ["-y", "x402-weblens-mcp"],
      "env": {
        "WEBLENS_WALLET_PRIVATE_KEY": "0x...",
        "WEBLENS_MAX_SPEND_USD": "1"
      }
    }
  }
}

For Claude Code: claude mcp add weblens -e WEBLENS_WALLET_PRIVATE_KEY=0x... -- npx -y x402-weblens-mcp

Related MCP server: SchemaFetch MCP Server

Settings

Variable

Default

Meaning

WEBLENS_WALLET_PRIVATE_KEY

required

Key of the wallet that pays. It stays on your computer and is only used to sign each payment.

WEBLENS_MAX_SPEND_USD

1

The most the server may spend in one session; further payments are refused.

WEBLENS_NETWORK

eip155:8453

Base mainnet. Use eip155:84532 for Base Sepolia testnet.

WEBLENS_API_URL

WebLens production

API address.

Safety: no single payment may exceed $0.05. Failed requests (status 400 or above) are not charged.

HTTP API and docs: https://x402-weblens-api-production.up.railway.app

Available Tools

10 tools
rugprobe_checkRugProbe token safety check ($0.01)A

Check an ERC-20 token on Base before trading: simulated buy and sell (honeypot, real buy/sell tax), liquidity, LP burn, owner powers (mint, blacklist, fees, pause), upgradeable proxy, and a 0-100 risk score. Automated on-chain heuristics, not financial advice. Costs $0.01 USDC.

Inputs: address (required).

rugprobe_gasBase gas advisor ($0.002)A

Should I submit a Base transaction now or wait? Returns next base fee, block fullness, slow/normal/fast EIP-1559 fee suggestions and the estimated USD cost of a transfer and a swap. Costs $0.002 USDC.

Inputs: no named parameters.

rugprobe_position_sizePosition size & ATR stop ($0.002)A

Position size and stop-loss from account risk: give price bars (high/low/close, oldest first) or an ATR plus entry, the account size and risk %. Returns Wilder ATR, stop price, units and position value. Math only, not financial advice. Costs $0.002 USDC.

Inputs: riskPct (required), accountSize (required), atr, bars, +4 more.

rugprobe_reportRugProbe Report: token due diligence ($0.05)A

Full due diligence on a Base token in one call: everything rugprobe_check does, plus the project's website (does it really list this contract? fake-site check, domain age, socials), a clear/caution/avoid verdict, a 0-100 risk score and a short summary. Automated risk signals, not financial advice. Costs $0.05 USDC.

Inputs: address (required), website.

weblens_csv_to_jsonCSV to typed JSON ($0.002)A

Convert CSV text to JSON rows with snake_case keys and a detected type per column (integer, number, boolean, date, datetime, string). Delimiter is auto-detected. Costs $0.002 USDC.

Inputs: csv (required), header, delimiter.

weblens_extractClean page text ($0.005)A

Get the main content of a web page as clean Markdown, without ads, menus or popups. Costs $0.005 USDC.

Inputs: url (required).

weblens_json_repairRepair & validate JSON ($0.002)A

Fix broken JSON (code fences, comments, single quotes, bare keys, trailing commas, Python literals, cut-off endings) and optionally validate it against a JSON Schema (no regex patterns). Costs $0.002 USDC.

Inputs: text (required), schema.

weblens_metadataPage metadata ($0.002)A

Get a web page's title, description, Open Graph tags, JSON-LD and favicon. Costs $0.002 USDC.

Inputs: url (required).

weblens_structureStructured data ($0.02)A

Turn a web page (url) or raw text into JSON that follows the JSON Schema you send. Costs $0.02 USDC.

Inputs: schema (required), url, text, instructions.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 10 tools

    Disambiguation4/5

    Most tools have clearly distinct purposes, and the two family prefixes (weblens_ vs rugprobe_) help separate web-content utilities from Base token risk tools. There is some overlap between rugprobe_check and rugprobe_report because the report is a superset, and weblens_extract vs weblens_structure could occasionally be confused, but the descriptions clarify when to use each.

    Naming Consistency5/5

    All tools follow a consistent snake_case family_action pattern, with weblens_ and rugprobe_ prefixes consistently identifying the product area. No mixed casing or vague verb styles appear.

    Tool Count5/5

    Ten tools is well within a reasonable range, and each tool maps to a distinct operation or analysis workflow. Neither family feels padded or thin.

    Completeness4/5

    The surface covers solid core workflows for both web extraction/conversion and Base token due diligence, including content extraction, metadata, links, schema structuring, token risk checks, position sizing, and gas estimation. Minor gaps exist such as no search/crawl/batch web operations or token history/portfolio tools, but these are not fatal for the stated utility scope.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Pay-per-use clean web reader for AI agents. URL in, markdown plus metadata out, in milliseconds. Settled per-call in USDC over x402 — no signup, no API keys.
      1
      102 npm
      3
      MIT
    • F
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents to extract structured JSON data from web pages using presets or custom JSON schemas, with confidence scores and source snippets, paid per call via USDC on Base.
      -
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables agents to convert any URL into clean, structured, agent-optimized content with a single paid call, paying $0.02 USDC per successful extraction via x402 on Base without API keys or accounts.
      MIT