{"id":"62cfe7f3-35ca-46c2-af6c-d817836dcff5","slug":"siteglass-mcp","kind":"mcp","name":"siteglass-mcp","summary":"Archive any URL to a shareable snapshot (rrweb/pdf/png) + agent-driven web QA — an MCP server for siteglass.io. No CAPTCHA, no signup.","owner":"siteglassio","repoUrl":"https://github.com/siteglassio/siteglass-mcp","license":"MIT","licenseTier":"permissive","trust":"community","stars":0,"match":null,"fewStars":true,"format":"mcp","createdAt":"2026-07-10T22:43:06.000Z","addedAt":"2026-10-14T06:11:07.429Z","logoUrl":"https://avatars.githubusercontent.com/u/302276521?v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["cloud-infra","browser-web","files-documents"],"tags":["ai-agent","archive","link-rot","playwright","rrweb","web-archiving","web-qa","website-screenshot"],"runtime":"node","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-14T19:43:19.932Z","sample":false,"url":"https://conduits.dev/tools/siteglass-mcp","language":"JavaScript","transports":["stdio"],"install":{"npm":"siteglass-mcp"},"requirements":{"env":[{"name":"SITEGLASS_API_KEY","secret":true,"required":false,"description":"siteglass API key (optional; the server auto-creates an account if absent)."},{"name":"SITEGLASS_BASE_URL","secret":false,"required":false,"description":"API base URL (default https://siteglass.io)."}],"upstreamMayCharge":false},"tools":null,"readme":"# siteglass MCP server\n\n[`siteglass-mcp`](https://www.npmjs.com/package/siteglass-mcp) exposes\n[siteglass.io](https://siteglass.io) as MCP tools. Three things an AI agent (or a\nhuman) can do with it:\n\n1. **Scan the app you're building — FREE, no account** — `siteglass_scan_local`\n   runs a headless browser *on your machine* against `http://localhost:PORT` (or\n   any URL you can reach — nothing is tunneled to us), then siteglass analyzes the\n   crawl and returns real findings: broken links/assets, console errors,\n   accessibility & SEO issues, forms worth testing, plus a plain-English report.\n   Point it at your dev server right after the agent builds or changes the app.\n   No signup, no ownership verification, no payment. (First run downloads a\n   headless Chromium, ~1 min.)\n2. **Archive any public URL** — `siteglass_render` renders a page server-side\n   and returns a permanent, shareable snapshot: an interactive **rrweb** DOM\n   capture (post-JS, selectable text, working links), a single-page **PDF**, or\n   a full-page **PNG**. No CAPTCHA, no signup, no ownership check, callable\n   programmatically — a bot-friendly alternative to archive.today. (It renders\n   JS-heavy pages; it does *not* produce WARC, and if the target itself blocks\n   datacenter IPs the capture can come back empty.)\n3. **Web-QA a deployed site you own** — **register → verify → scan → generate\n   flows → run → read results**, autonomously, no human signup. This is the\n   hosted/continuous tier (runs on our servers against your live site).\n\nThe **local scan (1) is free and needs no auth.** For the hosted tools (2, 3),\nauth is automatic (creates + caches an API key at `~/.siteglass/key`; set\n`SITEGLASS_API_KEY` to use your own); the first scan per site is free, then scans\n(2 credits) and flow runs (1 credit) draw from a prepaid balance topped up in USD\nover Bitcoin Lightning.\n\n## Install\n\n```sh\nnpx siteglass-mcp\n```\n\n**Claude Code:**\n```sh\nclaude mcp add siteglass -- npx siteglass-mcp\n```\n\n**Cursor / Claude Desktop / Windsurf** — add to the MCP config\n(`~/.cursor/mcp.json`, `claude_desktop_config.json`, …):\n```json\n{\n  \"mcpServers\": {\n    \"siteglass\": { \"command\": \"npx\", \"args\": [\"-y\", \"siteglass-mcp\"] }\n  }\n}\n```\n\n**OpenClaw** — add to `~/.openclaw/openclaw.json` (OpenClaw consumes MCP\nservers); your assistant gets the `siteglass_*` tools:\n```json\n{ \"mcpServers\": { \"siteglass\": { \"command\": \"npx\", \"args\": [\"-y\", \"siteglass-mcp\"] } } }\n```\n\nThen ask the agent: *\"test my app at <url> with siteglass.\"*\n\n## Tools\n\n| tool | does |\n|------|------|\n| `siteglass_scan_local` | **free, no-auth** — crawl a local/dev app in a browser on your machine → analyzed findings + report |\n| `siteglass_render` | archive any public URL → shareable snapshot permalink (rrweb / pdf / png); no CAPTCHA, no signup |\n| `siteglass_get_render` | fetch an archive once its Lightning invoice is paid |\n| `siteglass_register_site` | register an app URL, get an ownership-proof token |\n| `siteglass_verify_site` | confirm the token is live (DNS TXT or hosted) |\n| `siteglass_scan` / `siteglass_get_scan` | crawl the site → report + findings (first scan free) |\n| `siteglass_generate_flows` | derive executable E2E test flows from a scan |\n| `siteglass_run_flow` / `siteglass_get_run` | run a flow → per-step pass/fail, screenshots, video, rrweb replay |\n| `siteglass_set_credentials` | give a test login for authenticated flows |\n| `siteglass_feedback` | send feedback to the siteglass team |\n\nLong ops return an id and you poll the `get_*` tool, so every call stays\nunder MCP client request timeouts.\n\n## Discovery\n\n- Listed in the **official MCP Registry** as `io.siteglass/mcp`.\n- Machine-discovery on the API host: `https://siteglass.io/openapi.json`,\n  `/.well-known/agent-skills/index.json`, `/llms.txt`, `/SKILLS.md`.\n\nMIT licensed.\n"}