Start with a reproducible failure
A policy must distinguish intended scripts and resources from unwanted execution without breaking inquiries. Copying a header does not identify the website’s real dependencies.
Use a small controlled test to separate a symptom from its cause. Record the starting state, affected URL and observed outcome so that another person can repeat the check.
Collect evidence before changing anything
Compare Content-Security-Policy-Report-Only with the enforcing header on the same response. Report-only does not block. Tie each violation to a directive, resource and visitor action; browser extensions can create irrelevant noise.
Fictional example: A quote form uses an external check. The homepage looks correct, but submission stops because connect-src excludes the check’s request. Test the complete form journey with test data.
| Check | Expected outcome |
|---|---|
| Starting condition | An intended resource is blocked only during a particular action. |
| Implementation | Compare Content-Security-Policy-Report-Only with the enforcing header on the same response. Report-only does not block. Tie each violation to a directive, resource and visitor action; browser extensions can create irrelevant noise. |
| Verification | The same inquiry task works under enforcement without the exception admitting unexpected resources. |
Fix the cause and repeat the test
Correct the relevant directive or resource model. Do not broadly allow inline code to hide one isolated failure.
Repeat the same test after the change and add a nearby case that already worked. One successful case establishes that task in those conditions; it does not certify the whole website.
What the test establishes
The same inquiry task works under enforcement without the exception admitting unexpected resources.