# Agent Token Stripper

> Agent Token Stripper is a paid MCP server for web search and scraping. MCP server that strips HTML and scripts from web pages to cut LLM token usage. It is available on API and CLI.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, CLI
- **Repository:** https://github.com/goldminex/goldmindmax
- **MCP Registry name:** io.github.GoldMineX/agent-token-stripper
- **Categories:** [web search and scraping](https://ai-powered.com/categories/web-search-and-scraping)
- **Listing updated:** 2026-11-18

## About

The README describes this as an MCP tool that takes raw webpages and removes scripts, styles, navigation and header markup, returning plain text intended for an LLM context window. It positions the goal as lowering API token costs, though the README's savings figures (up to 80% and 15,000+ tokens per page) are the author's own claims and are not independently verified here.

According to the README, the server is paid per request via x402 micro-payments of $0.002 USDC on Base L2, so an agent can pay without a card or API key. It also sets a browser-style User-Agent and aborts target fetches after 8 seconds. The README shows adding it to Claude Code over HTTP.

It suits developers building agent pipelines that fetch many web pages and want smaller, cleaner inputs for the model.

## Features

- **HTML boilerplate stripping:** Removes script, style, nav and header elements and HTML tags from fetched pages, leaving plain text.
- **Per-request USDC payments:** Uses x402 so agents pay $0.002 USDC per request on Base L2 without API keys.
- **Browser-style User-Agent:** Sends browser-like headers to reduce 403 blocks on target URLs, according to the README.
- **8-second fetch timeout:** Aborts target page fetches after eight seconds to help keep agent pipelines from stalling.

## Good for

- Feeding cleaned web page text into an LLM context window
- Reducing token spend for agents that read many web pages
- Letting autonomous agents pay for web extraction on-chain
- Fetching pages that block default scraper user agents

## FAQ

### How much does a request cost?

The README states $0.002 USDC per request, settled via x402 on Base L2.

### How do I add it to Claude Code?

The README gives a single command: claude mcp add with the HTTP transport pointing at the hosted /api/extract endpoint.

### What does it remove from pages?

The README lists script, style, nav and header elements along with HTML tags.

### Is Agent Token Stripper free?

No. Agent Token Stripper is a paid product.

### What are the best alternatives to Agent Token Stripper?

The closest alternatives to Agent Token Stripper are Web Content Enrichment & Table Extractor (x402), AgentReader, Agent Read Proxy, AgentScraper API and Webshift.

## Where this comes from

- Description: written from the project's README and its MCP Registry entry.

## Alternatives

- [Web Content Enrichment & Table Extractor (x402)](https://ai-powered.com/mcp/web-content-enrichment-and-table-extractor-x402): Web Content Enrichment & Table Extractor (x402) is a paid MCP server for web search and scraping.
- [AgentReader](https://ai-powered.com/mcp/agentreader): AgentReader is a freemium MCP server for files and documents.
- [Agent Read Proxy](https://ai-powered.com/mcp/agent-read-proxy): Agent Read Proxy is a paid MCP server for web search and scraping.
- [AgentScraper API](https://ai-powered.com/mcp/agentscraper-api): AgentScraper API is a freemium MCP server for web search and scraping.
- [Webshift](https://ai-powered.com/mcp/webshift-3): Webshift is an open-source MCP server for web search and scraping.
- [Web Content Extract Mcp](https://ai-powered.com/mcp/web-content-extract-mcp): Web Content Extract Mcp is a freemium MCP server for web search and scraping.
- [URLpipe](https://ai-powered.com/mcp/urlpipe): URLpipe is a freemium MCP server for web search and scraping.
- [MTScraper](https://ai-powered.com/mcp/mtscraper): MTScraper is an open-source MCP server for web search and scraping.

---
Source: https://ai-powered.com/mcp/agent-token-stripper
