Get started
Introduction
What WebSecure is, what it checks and how to start.
WebSecure checks a public website for security problems and writes a plain-English report with the exact fix for each one. You can read it yourself, or hand it to Claude Code, Codex or Cursor and let the agent fix your code, then rescan to confirm.
Let your agent do it
Install the WebSecure skill and ask your agent to “scan my site with websecure and fix what it finds”.
mkdir -p ~/.claude/skills/websecure curl -fsSL https://scan.hoopgram.com/skill/SKILL.md -o ~/.claude/skills/websecure/SKILL.md
Details in Agent skill.
Why WebSecure
- One box, one minute. Paste a URL. A normal scan usually finishes in about a minute.
- Fixes, not just alerts. Every finding comes with the evidence, why it matters, and the config lines or code that fix it.
- Built for coding agents. The report ends with a ready prompt for your agent, and the API and skill let agents scan and rescan on their own.
- Honest results. Anything we could not check is listed as “not checked”, never counted as a pass. No “100% secure” badges.
- Safe by design. Anyone can run a normal scan, which only sees what a visitor sees. Deeper probing runs only on domains you prove you own.
- Free to use. Scans are free, with no sign-up. Protect watches a site every day for $9 a month. Pricing
Explore the docs
QuickstartRun your first scan and fix the top issue.
How a scan worksFrom URL to report, step by step.
Normal vs deep scanWhat each one checks and who can run it.
Verify your domainUnlock deep scans with a DNS record or a file.
Reading the reportSeverities, evidence, fixes and rescans.
API referenceStart scans and fetch reports from code.
What WebSecure does not do
It checks what the outside world can see. It does not log in to your site, test your business logic or permissions, or replace a human penetration test. The report says so at the end, every time.