How to fix: CSP Uses unsafe-hashes
CSP policy includes the unsafe-hashes directive, a hash-locked allowlist for specific pre-approved inline event handlers.
Why it matters
Unlike unsafe-inline, unsafe-hashes only permits inline event handlers whose exact content matches one of the policy's enumerated SHA hashes. An attacker who injects new or different markup cannot make it execute, since the hash of injected content won't match. Still worth cleaning up, but it does not enable arbitrary inline script execution.
unsafe-hashes is a deliberate hardening step (recommended by OWASP's CSP cheat sheet) for migrating legacy inline event handlers away from unsafe-inline while a full nonce-based refactor is in progress.
How to fix it
- 1Migrate inline handlers to external scripts with nonces.
Code examples
Replace unsafe-hashes with nonces
// unsafe-hashes allows inline event handlers with specific hashes, still risky
// Better approach: move all inline handlers to external scripts with nonces
import { randomBytes } from "crypto";
const nonce = randomBytes(16).toString("base64");
res.setHeader("Content-Security-Policy", `script-src 'nonce-${nonce}'; object-src 'none';`);
// Then add nonce={nonce} to your <script> tags onlyReferences
Detected automatically by VulnRadar. The check id is csp-unsafe-hashes, stable across every scan so you can filter or diff on it.
More headers checks
Questions
- How severe is CSP Uses unsafe-hashes?
- VulnRadar rates this medium severity in the headers category. The severity is fixed per check, not scored per site, so the same finding is the same severity on every scan.
- Does VulnRadar check for this automatically?
- Yes. It runs as part of every scan under the check id csp-unsafe-hashes, which stays the same between runs so you can filter, diff, or gate CI on it.
- Which standards does CSP Uses unsafe-hashes map to?
- It maps to CWE-693. In the OWASP Top 10 it falls under A05:2021.
Missing a check?
The detection set is open. If you know a misconfiguration this does not catch, propose it: the issue template asks for the threat model, how to detect it, the false-positive risk, and a test fixture, which is the same thing a maintainer would have to work out anyway.