From uptimerobot
Creates a free UptimeRobot HTTPS monitor using only the owner's email—no account, API key, or MCP auth. Useful when a user wants uptime monitoring but isn't signed in to UptimeRobot.
How this skill is triggered — by the user, by Claude, or both
Slash command
/uptimerobot:quick-monitor-setupThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **No MCP preflight needed.** Unlike the other UptimeRobot skills, this one does NOT use the `uptimerobot:*` MCP tools and does NOT require the user to authenticate. It is the path for when the user has no UptimeRobot account or API key. If the user IS already connected via the MCP server, prefer `create-http-monitor` instead.
No MCP preflight needed. Unlike the other UptimeRobot skills, this one does NOT use the
uptimerobot:*MCP tools and does NOT require the user to authenticate. It is the path for when the user has no UptimeRobot account or API key. If the user IS already connected via the MCP server, prefercreate-http-monitorinstead.
Use this when a user says something like "monitor my site and email me if it goes down" but has no UptimeRobot account or API key. You (the agent) do steps 1-4; the site owner does step 5 from their inbox. The result is the standard free plan: 50 monitors, 5-minute checks, free forever, no credit card.
The only thing the owner must provide is the email where down and up alerts go. Use an inbox they actually check.
https://api.uptimerobot.com
GET /agentic/agent-monitor/challenge?email=<owner-email>&url=<site>
url: the website to monitor. A bare domain works (www.example.com); https:// is assumed if you omit the scheme, so example.com is treated as https://example.com. Give an explicit http:// only if you specifically need plain HTTP.{ "nonce": "<hex>", "timestamp": <ms>, "difficulty": <bits>, "signature": "<hex>" }.difficulty from the response. Never hardcode it.Find a counter such that SHA-256("<nonce>|<counter>") has at least difficulty leading zero bits (pipe-separated, no spaces).
import hashlib
counter = 0
while True:
h = hashlib.sha256(f"{nonce}|{counter}".encode()).digest()
zeros = 0
for b in h:
if b == 0:
zeros += 8
else:
zeros += 8 - b.bit_length()
break
if zeros >= difficulty:
break
counter += 1
POST /agentic/agent-monitor
Content-Type: application/json
{
"email": "<owner-email>",
"url": "<same url you sent in step 1>",
"nonce": "<from challenge>",
"timestamp": <from challenge>,
"counter": <your solution>,
"signature": "<from challenge>"
}
Send the same url you used in step 1 (the signature is bound to it). Most responses return a uniform HTTP 200 { "status": "ok", ... } whether accepted or silently rejected (this is deliberate, it prevents probing who has an account). The one exception: an invalid proof of work returns HTTP 400 { "code": "INVALID_PROOF_OF_WORK" }, re-request a fresh challenge and solve again.
Tell the owner: "Check your inbox for an activation email from UptimeRobot and click the link to activate your monitor." No monitor exists until they do.
The owner clicks the activation link, lands on a confirmation page, and presses Activate. Only then are the account and monitor created. The link expires in 24h.
| Status | Meaning |
|---|---|
200 on submit | Accepted, or silently rejected (same body). Tell the user to check their inbox. |
400 INVALID_PROOF_OF_WORK on submit | Re-request a challenge and re-solve. |
400 INVALID_URL on challenge | The url was not a usable web address (empty or a non-http scheme). Fix it and retry. |
| activation link | The browser GET shows a confirm page for a valid token; the POST (the owner's click) creates everything. |
create-http-monitor: use this instead when the user IS connected via the UptimeRobot MCP server (it creates any monitor type on their existing account).setup: connect and authenticate the MCP server if the user wants the full toolset.npx claudepluginhub uptimerobot/aiCreates HTTP monitors in UptimeRobot to check URL reachability and response codes. Supports intervals, timeouts, auth, custom headers, SSL checks, and alert contacts.
Manages Better Stack uptime monitors: lists, creates, updates, pauses, deletes monitors, heartbeats, groups, and check types via API calls.
Automates Uptimerobot operations via Composio toolkit through Rube MCP. Discovers tool schemas, manages connections, and executes monitor/alert workflows.