Lyrenth
OfficialThis server turns public web pages into clean, low-token Markdown (AIDocument) you can read, batch-read, and meter through Lyrenth.
Read a single page (
read_url): fetch any public http(s) URL as a clean AIDocument — Markdown plus title, description, and structure, with navigation and boilerplate stripped; it reports found tokens and savings vs raw HTML.Control freshness (
fresh): force a new fetch instead of the cross-caller cached copy.Fit your context budget (
max_tokens): cap the returned body, trimmed at a clean paragraph or sentence boundary.Read many pages at once (
read_urls): batch 1–20 URLs in one call with per-URL error isolation, so one failure doesn't block the rest.Check your account (
check_usage): see plan tier, credits used vs monthly limit, credits remaining, and reset date.Safe, read-only operations: all tools are non-destructive, idempotent, and read-only;
read_url/read_urlsare open-world,check_usageis closed-world.
Lyrenth MCP server
Read the web through Lyrenth's index from any MCP client.
Exposes three tools:
read_urlturns a public web page into a clean AIDocument: stable Markdown plus title, description, and structure, with the navigation and boilerplate stripped. Your agent reads cleaned, low-token content instead of raw HTML, and every result shows how many tokens it saved vs the raw page.read_urlsdoes the same for up to 20 URLs in one batch call.check_usagereports your plan tier and credit usage.
Reads resolve through Lyrenth's cross-caller cache, and for verified domains they return the site's canonical version.
Every tool declares all four MCP tool annotations (read-only, non-destructive, idempotent, and open or closed world) as explicit booleans, and the HTTP transport ships with a test suite covering its security posture.
Hosted, no install
If your MCP client supports remote HTTP servers with a custom header (Claude and Cursor remote connectors do), there is nothing to run:
URL:
https://api.lyrenth.com/mcpHeader:
Authorization: Bearer aiwk_your_key_here
Same three tools, same key. The npx path below is for clients that speak MCP over stdio, and for anyone who prefers a local process.
Related MCP server: Exa MCP Server
Setup
Get a free API key at https://lyrenth.com/signup (2,000 reads/month, no card).
Add the server to your MCP client.
Claude Desktop (one click)
Download lyrenth-mcp.mcpb from the
latest release and
open it. Claude Desktop installs the server and asks for your API key; there is
no config file to edit. The bundle carries the same code as the npm package.
To build it yourself: ./scripts/build-mcpb.sh.
Claude Desktop / Cursor (manual config)
Add to your MCP config (Claude Desktop: claude_desktop_config.json):
{
"mcpServers": {
"lyrenth": {
"command": "npx",
"args": ["-y", "lyrenth-mcp"],
"env": { "LYRENTH_API_KEY": "aiwk_your_key_here" }
}
}
}Claude Code
claude mcp add lyrenth -e LYRENTH_API_KEY=aiwk_your_key_here -- npx -y lyrenth-mcpThen ask your assistant to read a page, for example: "Read
https://example.com/article and summarize it." It will call read_url and get
the cleaned AIDocument back.
Tools
Tool | Arguments | Returns |
|
| The page as a clean AIDocument: a short provenance header (token count + how much smaller than raw HTML) plus the Markdown body. |
|
| Up to 20 pages in one call, each as a clean AIDocument, with per-URL error isolation (a failed URL is reported and doesn't block the others). Billed one credit per successfully-read URL. |
| none | Your plan tier, credits used against your monthly limit, credits remaining, and the reset date. |
Configuration
Env var | Required | Default | Notes |
| yes | none | Free key at https://lyrenth.com/signup |
| no |
| Override for staging or self-host |
Why read through Lyrenth
Cleaner, cheaper. One stable AIDocument shape per URL; far fewer tokens than raw HTML to a model.
Cached across callers. The same URL fetched by many agents collapses to a minimal number of origin fetches, so it is fast and origin-friendly.
Fresh when verified. When a site's owner has verified with Lyrenth, a page they change is fetched again at the front of our crawl queue, so you read the new version sooner.
Privacy
The server sends exactly two things to Lyrenth's API (api.lyrenth.com): the URLs you ask it to read, and your API key to authenticate and meter the call. Nothing else leaves your machine: no page content you hold locally, no conversation context, no telemetry. How Lyrenth handles fetched pages and account data is covered by the privacy policy: https://www.lyrenth.com/privacy.
License
MIT. See LICENSE.
Local build
npm install
npm run build
LYRENTH_API_KEY=aiwk_... node dist/index.js # speaks MCP over stdioPart of the Lyrenth project. The AIDocument format is an open contract; see https://lyrenth.com/llms-full.txt.
Available Tools
3 toolscheck_usageCheck usageARead-onlyIdempotent
Check your Lyrenth credit usage: plan tier, credits used against your monthly limit, credits remaining, and the reset date. Takes no arguments.
Inputs: no named parameters.
read_urlRead URLARead-onlyIdempotent
Read any public web page as a clean AIDocument: Markdown plus title, description, and structure, with navigation and boilerplate stripped. Prefer this over a raw HTTP fetch whenever you need the content of a web page; it returns far cleaner, lower-token text. Powered by Lyrenth.
Inputs: url (required), fresh, max_tokens.
read_urlsRead URLs (batch)ARead-onlyIdempotent
Read several public web pages in one batch call, each as a clean AIDocument. Up to 20 URLs, faster than calling read_url repeatedly. Use it to compare or summarize multiple pages at once; a failed URL is reported per-item and does not block the others. Powered by Lyrenth.
Inputs: urls (required), fresh, max_tokens.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
read_url and read_urls are clearly distinguished as single-page versus batch reading, while check_usage serves a completely separate purpose. There is no meaningful overlap or ambiguity between any of the tools.
All tool names follow a consistent verb_noun snake_case pattern: read_url, read_urls, and check_usage. The pluralization of read_urls naturally signals the batch variant without breaking the naming convention.
Three tools is well-scoped for a focused web-reading server: single read, batch read, and usage checking each earn their place. The small surface area matches the narrow, clear purpose.
The tool set fully covers the apparent domain: reading a single page, reading multiple pages at once, and checking account usage. There are no obvious missing operations or dead ends for the stated purpose.
Maintenance
Related MCP Connectors
Read any public web page as clean Markdown for LLMs, with its metadata. Paid per call, x402.
Read any web page as clean Markdown for AI agents: fetch, search, metadata, links. SSRF-safe.
Read authorized public pages as Markdown or HTML metadata. Operated by an autonomous AI agent.
Read any public web page as clean Markdown for LLMs. Respects robots.txt. Free tier; x402 paid API.
Related MCP Servers
- AlicenseAqualityBmaintenanceThis server enables LLMs to retrieve and process content from web pages, converting HTML to markdown for easier consumption.101225,076 PyPI91,952MIT
- AlicenseAqualityBmaintenanceA Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.846,492 npm5,232MIT
- AlicenseBqualityDmaintenanceExposes the Exa web search API as an MCP server, providing tools for web searching, content retrieval, and similar page discovery. It also supports advanced features like AI-generated answers and automated research task management.9MIT
- AlicenseAqualityCmaintenanceEnables AI agents to read web pages reliably, returning clean markdown content, hyperlinks, and metadata without navigation or ad noise.34 npmMIT