Changelog

• Agent skill. Install with npx skills add ScaledPapers/skills --skill scaledpapers -a claude-code -a codex -g -y, then run bash ~/.agents/skills/scaledpapers/scripts/wire-mcp.sh for your editor (cursor, claude, or codex). Listed on skills.sh. See MCP and the setup prompt.

• Enterprise pricing. The homepage and pricing page add an Enterprise card at a custom price: everything in Pro, with check volume and terms agreed separately. Talk to Nurbol opens the demo scheduler.
• Free tools. A Free tools menu in the header and footer collects the website down checker, a new uptime calculator (SLA percentage to downtime per day, month, and year, and back), a status badge builder, and the GPU resource calculator, now hosted here. All four live under /tools and need no account.
• Status badge. Every public status page serves a live SVG at {slug}.status.scaledpapers.com/badge.svg showing its overall state, with label and style parameters (flat, flat-square, for-the-badge, landing). Landing is a colored dot and a sentence for a footer or marketing page. Cached five minutes; private pages return 404. See Status pages.
• New homepage. The landing page now tells the product story in order: live checks, the MCP server, how alerts confirm before they send, the dashboard, probe regions, custom domains, the incident lifecycle, setup snippets, pricing, and the demo video, all on one page. The free website down checker moved to /website-down-checker; old /#checker links redirect there.

• MCP. An owner or admin creates one API key and connects Cursor, Claude Code, or Codex. The agent can set up checks, a status page, incidents, and maintenance windows. See MCP. An agent can follow the setup prompt to register the server and call usage.

• Subscribe to Pro. Settings Subscribe opens checkout for the selected workspace. Pro is billed by Lemon Squeezy. Manage billing opens the customer portal.
• Delete workspace. The workspace owner can delete the selected workspace from Settings. Type the confirm token (your email for Personal, or the workspace name) to confirm. This removes that workspace’s status pages, checks, API keys, alerts, and team members. A paid period that is still active, including one scheduled to cancel at period end, must end or be canceled in Manage billing first. Billing history stays on file.
• Delete account. If that workspace is your only owned workspace, Settings offers Delete account instead: the same workspace wipe, then your sign-in is removed and you are signed out.

• Page settings. The console screen formerly Page info is Page settings. Delete page is a danger zone at the bottom of that screen, not a sidebar item. /app/{slug}/delete still opens it. The Status pages row menu still has Delete page.
• Private status pages. On Pro, check Make status page private when you create a page or later on Page settings. Visitors enter a shared password once to view it. Subscribe and RSS stay off so the feed cannot be shared without the password. The Status Pages API accepts visibility: "internal" and password on POST /v1/sites and PATCH /v1/sites/{slug}. Free pages stay public.
• Custom domain. On Pro, set status.yourcompany.com when you create a page or on Page settings. CNAME the hostname to cname.scaledpapers.com, add the TXT challenge, and Verify DNS. The public URL switches when TLS is live. Private pages can use the same host.

• Status page tab icon. A saved PNG or SVG company logo replaces the ScaledPapers favicon on that public page.

• Page info. Website, support, and privacy URLs are labeled Optional. If you set them, they appear on the public status page (page name link, Support, subscribe privacy). Company logo is optional and, if set, shows next to the page name. Page time zone stays an owner setting for dates, not a public field. Organization legal name is no longer on this form (invoicing comes later with checkout).

• Apps. Status page Apps lists Slack as Coming soon. Visitor Slack subscribe (/feed subscribe) is unchanged.

• Settings. Open Settings for plan usage of the selected workspace (checks, monthly probes, status pages, region cap) and your account email. Paid checkout is not live; View plans goes to Pricing.
• Maintenance windows. Schedule a window from Maintenance. It pauses the selected checks and marks targeted status page components Under maintenance. The public page lists Affected components and remaining duration. Notify subscribers emails status-page subscribers when a public component is included.

• Team workspaces. Invite teammates from Team. Owner, admin, and member roles share checks and status pages. Owner and admin create API keys for the selected workspace; members cannot. A key stays valid if that admin is later changed to member until someone revokes it.
• DNS checks. Add a DNS check from Checks → DNS or POST /v1/endpoints with type: "dns". Set the domain, record type (A, AAAA, CNAME, TXT, MX, NS, CAA), UDP or TCP, an optional resolver, and expected values. Probes run from the same five Americas regions as HTTP checks. A successful lookup is up; mismatched records are down.
• TCP checks. Add a TCP check from Checks → TCP or type: "tcp". Set the host, port, IP family, optional payload, and expected outcome: Connection should succeed or Connection should not succeed. Samples include connect time and the resolved IP. Private and loopback targets are rejected.

• Status Pages API. Manage status pages with the same API key as the Checks API at https://api.scaledpapers.com: create, list, and delete pages, add components, and open incidents. Incidents accept component_ids with component public ids (the same 12-character ids as console URLs).
• Incident updates via API. Post investigation updates and resolve incidents with POST /v1/sites/{slug}/incidents/{public_id}/updates. Status resolved closes the incident; subscribers and Slack are notified as from the console.

• Landing page. The homepage leads with status pages and multi-region monitoring. Product sections show live uptime and status previews. Website down checker scrolls cleanly to the checker, and after a check you can open an uptime monitoring check.
• Console on mobile. Open navigation from a hamburger next to the ScaledPapers wordmark. The brand mark hides on narrow widths so the header stays clear.

• Check email alerts. Down and recovery emails are clearer: plain text with region detail and links back to the check.
• Status incidents from checks. When a linked check opens an incident, the title uses the status component name.

• System metrics. Public status pages show latency charts in their own panel. Hover for time and value; filter by AWS region code. Owner Ping matches.
• API keys. Create a key in the console and call https://api.scaledpapers.com as Authorization: Bearer sp_… to list, add, pause, or run your checks. The secret is shown once.
• Link a check to a status page. On the check form, pick a page and component. Optionally open an incident when the check is down, and show the metric on the public page.

• API and Website checks. Add a check from the console, see live status and a 24-hour uptime bar, and open a detail page with a results chart and recent samples. Run, pause, or edit from there. Type a hostname like example.com — https is added for you. Status updates on its own.

First public release.

• Status pages. Create a public status page from the console: components, incidents, and 90-day uptime. Visitors can subscribe by email, RSS, or Slack. Live pages use {slug}.status.scaledpapers.com.
• Website down checker. Check a site you are allowed to test from 16 regions and see HTTP status, latency, and where it is down.
• Daily research papers. A day-based stack to skip, save, read, take notes, and ask AI about a paper.