SECURITY TOOLS
CSP Generator & Evaluator
Build a starter Content Security Policy or review a raw policy for a focused set of explainable weaknesses—without assigning a misleading security score.
LOCAL POLICY ANALYSIS
Build or evaluate a policy
Paste the policy value or the full Content-Security-Policy header. Duplicate directives and selected risks are reported.
Generated output
Policy findings
Findings are intentionally limited; absence of a warning does not prove the policy is secure.
No findings from this limited rule set.
Your policy stays in this browser. Test in Report-Only mode before enforcing changes.
CSP generator and evaluator guide
A Content Security Policy lets a site restrict where browsers may load resources and execute code. Start with the policy your application actually serves, or use the builder for a draft that you will test and refine.
Use findings as prompts, not a verdict
The evaluator flags selected patterns such as 'unsafe-inline', 'unsafe-eval', broad wildcards, insecure HTTP sources, and missing object-src, base-uri or frame-ancestors directives. It does not model every CSP directive, browser version, application route, nonce flow, hash, or third-party dependency.
A nonce must be generated securely on the server for each response and match the nonce attribute on the allowed element. A hash must match the exact inline content. This tool does not create production nonces or hash an application's inline code. Never copy a generic policy into production without testing the site's required scripts, styles, images, connections, fonts and embeds.
Roll out with Report-Only first
Where supported, deploy Content-Security-Policy-Report-Only to observe violations without blocking the page. Review reports against actual application behavior, remove unnecessary sources, and only then consider an enforcing header. The builder's same-origin starter may break sites that depend on inline code, remote assets, analytics, APIs or embedded content.
Apache: Header set Content-Security-Policy-Report-Only "default-src 'self'; ..."
Nginx: add_header Content-Security-Policy-Report-Only "default-src 'self'; ..." always;These are illustrative snippets; configure your server module, quoting and policy value for your environment.
Frequently asked questions
Should I deploy the generated policy immediately?
No. Start with Content-Security-Policy-Report-Only where possible, inspect real violations, and adjust the allow-list before enforcing. A policy that is too strict can break scripts, styles, images, APIs, embeds or other site features.
Does the evaluator give a security score?
No. It reports a narrow set of explainable findings such as unsafe-inline, unsafe-eval, broad wildcards and missing defensive directives. It is not a complete security audit or proof that a policy is safe.
How should I use nonces or hashes?
A nonce must be unpredictable and freshly generated for each response, then placed both in the CSP source list and on the matching script or style element. A hash must match the exact inline content. This browser-only tool does not generate production nonces or calculate hashes for your application.
Can I put every directive in a meta element?
No. Some policy features, including frame-ancestors and reporting directives, are not supported from a CSP meta element. The converter omits those directives and reports each omission; an HTTP response header is generally the more capable delivery method.
Is my policy uploaded?
No. Parsing, evaluation and policy generation run locally in your browser.
Sources
Security note: this is a focused policy helper, not a full evaluator, automated deployment, or security certification.