DechoNet

DechoNet checks domains, links and IP addresses from the outside, using only public information, for site owners, developers, security staff, people building AI agents and anyone who got a suspicious link. Seen. Remembered. Signed.

What you can ask Claude to do with it:

Give a domain an overall security score (0–100, grade A+ to F) with the fixes that matter most, across DNS, the TLS certificate, HTTP security headers, email authentication, registration data and network — passive, public lookups only. Check a page before logging in, paying, submitting a form or downloading: phishing signals, where card fields are hosted, and hidden text addressed to AI agents. Every answer lists what was not checked. Check whether a link from a text message or email shows phishing signals. Redirects and shorteners are followed to the real destination. Write down what should stay true about your domain (HTTPS works, a header is present, DMARC policy, certificate days left) and have DechoNet confirm it from the outside, now and daily, with where to fix what does not hold. Run a single check for any one area: DNS, certificate and TLS, email authentication, response headers, registration, network, subdomains, look-alike domains. See what changed on a domain, and report a page or MCP server that looks malicious so DechoNet checks it itself. How it works: checks run from Cloudflare's global edge plus a vantage point in Seoul. Domain, host and IP results come with a signed observation record that anyone can verify against DechoNet's public key.

Limits: DechoNet reads public data only. Ports are checked only on request. It does not log in or test apps from the inside. A result without warning signs is not a guarantee; it is a snapshot at the time shown. A report alone never changes a verdict. Tools that save something (monitoring, expectations, reports) are marked as such in their annotations; all others only read.