CSP Evaluator
Check a Content Security Policy for common weaknesses. Paste a policy, a Content-Security-Policy header line or a meta tag and the evaluator lists every directive and finding by severity: scripts allowed through unsafe-inline, unsafe-eval, wildcards or whole schemes, missing object-src, base-uri and frame-ancestors, unquoted keywords, short nonces, deprecated and unknown directives.
Frequently asked questions
What counts as a high severity finding?
Anything that lets an attacker run scripts: no script-src or default-src at all, unsafe-inline without a nonce or hash, a wildcard or a whole scheme such as https: or data: in the script sources, a missing object-src, and keywords such as self written without quotes.
Why is unsafe-inline sometimes only information?
When script-src also contains a nonce or a hash, browsers that support them ignore unsafe-inline. It is then only a fallback for very old browsers and does not weaken the policy in current ones.
How is strict-dynamic treated?
With strict-dynamic, browsers ignore host and scheme sources in script-src, so the evaluator does not flag them. strict-dynamic without a nonce or hash is reported, because then no script can start the chain of trust.
Which input formats are accepted?
A bare policy, a full header line starting with Content-Security-Policy: or Content-Security-Policy-Report-Only:, or a meta tag with http-equiv="Content-Security-Policy". Several policies separated by commas are evaluated one by one.
Does it fetch the policy of a website?
No. It only evaluates the text you paste. Copy the header from your browser's developer tools or from a header checker first.
Does a clean result mean my site is safe?
No. The evaluator checks the policy text against known weaknesses. Allowed domains can still host scripts an attacker could use, and a policy does not replace escaping and validating input in your code.