check_watch
Poll a Skim Watch for content changes (or fetch registration status). Pass the watch_id from watch_urls. Card lane: GET /api/t/watch/diff?id= (or /status). Wallet lane: GET /api/v2/watch/diff?id= (or /status, which is free).
LABSCO SUMMARY
Page-to-Markdown with a batch path and a change watcher, not just a fetcher.
The output is a quarter the size of the raw HTML, which is the difference between reading five pages and reading twenty in the same context. extract_url is the sharper tool: it fills a schema you define from what is actually on the page and leaves missing fields missing, rather than letting a model fill the gaps.
What it is
The official MCP server for Skim. It fetches a URL and returns the article as Markdown with a YAML frontmatter block — no navigation, ads or boilerplate — and can pull structured JSON out of a page against a schema or a preset. It also registers watches that report content diffs on later checks.
What you get
- read_url — one URL as clean Markdown with title, byline, published date, language and excerpt in frontmatter; roughly a quarter the size of the raw HTML
- read_urls — up to ten URLs in one call, returning per-URL Markdown and metadata, instead of looping read_url
- extract_url — structured JSON or tables from a page against a JSON Schema you supply, or a preset: article, product, job, review, event or table; values come only from the page
- watch_urls — register a private watch on 1 to 20 URLs and get back a watch_id to keep secret
- check_watch — poll a watch; the first check baselines each page and later ones report the content diff
Requirements
A free Skim API key (it starts with sk402_) from skim402.com, set as SKIM_API_KEY; the free plan allows 1,000 reads a month and a card is required at signup. Clients run npx -y skim-mcp over stdio. An optional wallet path replaces the key entirely: set SKIM_WALLET_PRIVATE_KEY and calls pay per request in USDC on Base at $0.002 each — use a dedicated wallet, since the key sits in a plaintext config file.
Setup effort
One command plus a key — npx -y skim-mcp, then supply credentials
USE CASES
01
Reading an article without the page furniture
Fetch the URL as Markdown with its title, byline and date already parsed out.
02
Researching across several sources at once
Send up to ten URLs in a single call rather than fetching them one at a time.
03
Pulling fields out of a listing page
Give a schema or pick a preset and get back JSON where every value came from the page.
04
Noticing when a page changes
Register a watch over the pages you care about, then poll it for the content diff.
5 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
extract_url
Extract structured JSON (or tables) from a URL via Skim. Pass a JSON Schema, or a preset (article, product, job, review, event, table). Values come only from the page — missing fields are omitted or null, never invented. Card lane: POST /api/t/extract. Wallet lane: POST /api/v1/extract.
read_url
Fetch any URL and return clean, agent-ready Markdown via Skim (skim402.com). Output is ~4x smaller than raw HTML — fewer tokens, faster processing. Strips nav, ads, and boilerplate; preserves the article body plus structured metadata (title, byline, published date, language, excerpt). Use this whenever you need to read web content: articles, docs, blog posts, GitHub READMEs, research papers, etc.
read_urls
Read up to 10 URLs in one Skim call and return per-URL clean Markdown plus metadata. Use this instead of calling read_url in a loop. Card lane: POST /api/t/read/batch. Wallet lane: POST /api/v1/read/batch.
watch_urls
Register a private Skim Watch on 1–20 URLs. Returns a watch_id (treat it as a secret) used with check_watch. First check baselines each page; later checks report content diffs. Card lane: POST /api/t/watch (intended token path from skim402-web; may 404 until that route is live). Wallet lane: POST /api/v2/watch.
BEFORE YOU INSTALL
What to know first
Tools you get
5 tools when we connected; the project documents 8. crawl_url, read_pdf and poll_signal need SKIM_API_KEY; with only a wallet configured, you get the other 5.
Requires
- npx on PATH
- SKIM_API_KEY in your environment
Where it runs
All 4 documented clients run the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
npx -y skim-mcpMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"skim": {
"command": "npx",
"args": [
"-y",
"skim-mcp"
],
"env": {
"SKIM_API_KEY": "<YOUR_SKIM_API_KEY>"
}
}
}
}~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)
Cursor
CURSOR · JSON
{
"mcpServers": {
"skim": {
"command": "npx",
"args": [
"-y",
"skim-mcp"
],
"env": {
"SKIM_API_KEY": "<YOUR_SKIM_API_KEY>"
}
}
}
}~/.cursor/mcp.json (or Settings → MCP)
Cline, Continue
JSON
{
"mcpServers": {
"skim": {
"command": "npx",
"args": [
"-y",
"skim-mcp"
],
"env": {
"SKIM_API_KEY": "<YOUR_SKIM_API_KEY>"
}
}
}
}
