Siteglass

io.siteglassPublished by siteglass.iov0.1.5Updated Oct 24, 2026

Archive any URL to a shareable rrweb/pdf/png snapshot + agent-driven web QA (scan, run E2E flows).

VerifiedSTDIODesktop onlyWeb Search & ScrapingBrowser AutomationDeveloper Tools

Overview

AI-generated overview

Lets an assistant scan a local or deployed web app for issues, archive public URLs as shareable snapshots, and generate and run E2E test flows.

What you can do
Exposes siteglass.io as MCP tools. siteglass_scan_local runs a headless browser on your machine against a local or reachable URL and returns findings such as broken links, console errors, accessibility and SEO issues, plus a report. siteglass_render archives any public URL as an rrweb DOM capture, PDF or PNG, and hosted tools register and verify a site, scan it, generate E2E flows, run them and fetch results. Long operations return an id that is polled through get_ tools.
When to use it
Useful right after an assistant builds or changes a web app and you want a quick free local QA scan, when you need a permanent shareable snapshot of a public page, or when you want agent-driven E2E testing of a deployed site you own.
Requirements
Runs locally via npx siteglass-mcp, so Node.js and network access are needed; the first local scan downloads a headless Chromium. The local scan needs no account. Hosted tools auto-create and cache an API key at ~/.siteglass/key, or use SITEGLASS_API_KEY; SITEGLASS_BASE_URL optionally overrides the API base URL.
Before you install
Hosted scans and flow runs consume prepaid credits topped up in USD over Bitcoin Lightning after the first free scan per site; siteglass_get_render fetches an archive only once its Lightning invoice is paid. siteglass_set_credentials supplies a test login for authenticated flows, so treat those credentials as sensitive. Registering and verifying a site requires an ownership-proof token via DNS TXT or hosted verification. Local scan traffic stays on your machine, but hosted tools send your site's

Installation

In SourceWeft

  1. Open Siteglass in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

siteglass MCP server

siteglass-mcp exposes siteglass.io as MCP tools. Three things an AI agent (or a human) can do with it:

  1. Scan the app you're building — FREE, no account — siteglass_scan_local runs a headless browser on your machine against http://localhost:PORT (or any URL you can reach — nothing is tunneled to us), then siteglass analyzes the crawl and returns real findings: broken links/assets, console errors, accessibility & SEO issues, forms worth testing, plus a plain-English report. Point it at your dev server right after the agent builds or changes the app. No signup, no ownership verification, no payment. (First run downloads a headless Chromium, ~1 min.)
  2. Archive any public URL — siteglass_render renders a page server-side and returns a permanent, shareable snapshot: an interactive rrweb DOM capture (post-JS, selectable text, working links), a single-page PDF, or a full-page PNG. No CAPTCHA, no signup, no ownership check, callable programmatically — a bot-friendly alternative to archive.today. (It renders JS-heavy pages; it does not produce WARC, and if the target itself blocks datacenter IPs the capture can come back empty.)
  3. Web-QA a deployed site you own — register → verify → scan → generate flows → run → read results, autonomously, no human signup. This is the hosted/continuous tier (runs on our servers against your live site).

The local scan (1) is free and needs no auth. For the hosted tools (2, 3), auth is automatic (creates + caches an API key at ~/.siteglass/key; set SITEGLASS_API_KEY to use your own); the first scan per site is free, then scans (2 credits) and flow runs (1 credit) draw from a prepaid balance topped up in USD over Bitcoin Lightning.

Install

sh
npx siteglass-mcp

Claude Code:

sh
claude mcp add siteglass -- npx siteglass-mcp

Cursor / Claude Desktop / Windsurf — add to the MCP config (~/.cursor/mcp.json, claude_desktop_config.json, …):

json
{  "mcpServers": {    "siteglass": { "command": "npx", "args": ["-y", "siteglass-mcp"] }  }}

OpenClaw — add to ~/.openclaw/openclaw.json (OpenClaw consumes MCP servers); your assistant gets the siteglass_* tools:

json
{ "mcpServers": { "siteglass": { "command": "npx", "args": ["-y", "siteglass-mcp"] } } }

Then ask the agent: "test my app at with siteglass."

Tools

tooldoes
siteglass_scan_localfree, no-auth — crawl a local/dev app in a browser on your machine → analyzed findings + report
siteglass_renderarchive any public URL → shareable snapshot permalink (rrweb / pdf / png); no CAPTCHA, no signup
siteglass_get_renderfetch an archive once its Lightning invoice is paid
siteglass_register_siteregister an app URL, get an ownership-proof token
siteglass_verify_siteconfirm the token is live (DNS TXT or hosted)
siteglass_scan / siteglass_get_scancrawl the site → report + findings (first scan free)
siteglass_generate_flowsderive executable E2E test flows from a scan
siteglass_run_flow / siteglass_get_runrun a flow → per-step pass/fail, screenshots, video, rrweb replay
siteglass_set_credentialsgive a test login for authenticated flows
siteglass_feedbacksend feedback to the siteglass team

Long ops return an id and you poll the get_* tool, so every call stays under MCP client request timeouts.

Discovery

  • Listed in the official MCP Registry as io.siteglass/mcp.
  • Machine-discovery on the API host: https://siteglass.io/openapi.json, /.well-known/agent-skills/index.json, /llms.txt, /SKILLS.md.

MIT licensed.

Source: README.md at commit 881c376

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.5LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts