The IsItDown.now skill for OpenClaw
Give an OpenClaw agent measured website status, confirmed outages, reporting spikes and 30-day uptime from the read-only IsItDown.now API. The skill tells the agent which endpoint to call, how to interpret each state and where its evidence comes from.
Install
Install the published skill into the active OpenClaw workspace with one command:
openclaw skills install @henrick86/isitdown-now
View on ClawHub
The package contains instructions only. It does not contain an API key, executable code or a background process.
What an agent can answer
- Whether a continuously monitored service has a confirmed outage right now.
- Which monitored services are down, slow or attracting an unusual number of reports.
- A service's latest response time, HTTP status, 30-day uptime and recent incident record.
- The difference between a confirmed outage, a failed check awaiting confirmation and public reports alone.
The REST API reads measurements IsItDown.now has already taken. It does not send a new request to a service when the agent calls it.
Get a free API key
- Sign in at the API-key dashboard and claim a domain you control.
- Verify the domain with a DNS TXT record, a meta tag or the small verification file.
- Create a key and copy it when shown. The plaintext key is displayed only once.
- Store it as a secret named ISITDOWN_API_KEY in the agent's runtime.
The overall summary works without a key. Other API routes require a bearer key. The free tier includes 1,000 requests per UTC day across the account.
Endpoints in the skill
| Endpoint | Use |
|---|---|
| GET /api/v1/summary | A key-free overview of checks, reports and confirmed outages. |
| GET /api/v1/status/<domain> | Current state, timing, uptime and incident details for one service. |
| GET /api/v1/services | Find the exact domain or slug for a continuously monitored service. |
| GET /api/v1/down | List every confirmed outage happening now. |
| GET /api/v1/trending | Find services with the most user reports in the past six hours. |
Read the complete API documentation, including response fields, limits and error codes.
Evidence before certainty
The skill requires the agent to name the measurement time and link to the returned status page. It also prevents a report spike from being presented as a confirmed outage: a continuously monitored outage is recorded only after consecutive failed checks.
A third-party measurement cannot diagnose the user's Wi-Fi, ISP or device. If the requested domain is not in the continuously monitored directory, the agent says so instead of treating an unknown service as down.