Uptime & reliability monitoring
Our agents watch your public sites. A private Go agent watches what's behind your firewall. You only hear from us when it's real.
No credit card · free plan forever
The gap
Your homepage can be green while the database behind it refuses every connection. The 4am pages usually don't have a public URL — CuliPulse watches those too.
How it works
Paste a URL and we check it from our shared locations in Singapore, Frankfurt and Montreal — from outside your network, so you see what your users see, including problems your own network never reveals. They can only reach public addresses: for a private host, use your own agent.
One static Go binary on a box inside your network runs the same checks and streams results out over an outbound 443 connection. No inbound ports, no VPN. Turn on auto-discovery and it suggests monitors for what it can reach — you approve what matters.
$ curl -fsSL https://culipulse.dev/install.sh \
| sudo CULIPULSE_API_URL=https://culipulse.dev \
CULIPULSE_AGENT_TOKEN=<your-token> sh
Audit the source on GitHub →
Checks
Every check runs from our shared agents or your own private agent — same alerting rules either way (private hosts need your own agent). Even the third-party vendors you depend on, watched from their own status page and pinned to the monitors that rely on them.
Vendor & dependencies
Half your 3am pages aren't your fault — an auth provider, a payment API, a cloud region. CuliPulse watches your vendors' own status and pins it to the monitors that depend on them, so when you open a red monitor you can already see whether one of its dependencies is having a bad day too.
A vendor's own status page — Anthropic, AWS by service & region, Cloudflare, GitHub and more — turned into a monitor you can alert on.
Pin the vendors a monitor leans on; their live status shows right on that monitor's page and rolls up into one line.
You see your monitor's state and each dependency's status side by side. CuliPulse doesn't claim the vendor caused it — the judgment stays yours.
Smart alerting
One flaky checkpoint isn't an outage. On Solo and Pro, CuliPulse can require more than one checkpoint to agree before it alerts you, so a single bad route or a two-second blip won't wake you up. If one location stops reporting, the others keep watching and can raise the alert on their own. While it lasts, the monitor's per-location results still show which checkpoint is failing.
Status & reports
Turn any set of monitors into a shareable SLA report or public status page — no login for the people you send it to.
Every credential, header and payload you store is encrypted at rest with authenticated encryption.
The agent dials out over 443. You never open a port or hand us network access.
Each agent gets its own key, scoped to its own checks and revocable on its own.
Pricing
Built on Cloudflare, so the free tier stays free. Paid plans buy volume, faster checks, more locations and longer history.
Public sites in a minute. Internal services in about five.
No credit card · free plan forever