CVE-2020-5217 is a medium-severity (CVSS 5.8) CWE-95 vulnerability in secure_headers. 1 public exploit reference exists, so weaponization risk is real. A fix is available for secure_headers — see the affected versions and patch details below.
Directive injection when using dynamic overrides with user input
EPSS Exploitation Probability
Probability of exploitation in the next 30 days, from FIRST.org EPSS.
How urgent is this, really
CVE-2020-5217 by exploitation likelihood (EPSS) against impact (CVSS). Outside the shaded patch-first corner.
Where this sits among everything scored
Of 394,803 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Counts from FIRST.org, log-scaled.
Real-World Exposure
secure_headers💎secure_headers💎secure_headersReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects RubyGems packages — download data is not available via public APIs for these ecosystems.
Description
Impact
If user-supplied input was passed into append/override_content_security_policy_directives, a semicolon could be injected leading to directive injection.
This could be used to e.g. override a script-src directive. Duplicate directives are ignored and the first one wins. The directives in secure_headers are sorted alphabetically so they pretty much all come before script-src. A previously undefined directive would receive a value even if SecureHeaders::OPT_OUT was supplied.
The fixed versions will silently convert the semicolons to spaces and emit a deprecation warning when this happens. This will result in innocuous browser console messages if being exploited/accidentally used. In future releases, we will raise application errors resulting in 500s.
Duplicate script-src directives detected. All but the first instance will be ignored.
See https://www.w3.org/TR/CSP3/#parse-serialized-policy
Note: In this case, the user agent SHOULD notify developers that a duplicate directive was ignored. A console warning might be appropriate, for example.
Patches
Depending on what major version you are using, the fixed versions are 6.2.0, 5.1.0, 3.8.0.
Workarounds
If you are passing user input into the above methods, you could filter out the input:
override_content_security_policy_directives(:frame_src, [user_input.gsub(";", " ")])
References
Reported in https://github.com/twitter/secure_headers/issues/418 https://www.w3.org/TR/CSP3/#parse-serialized-policy
For more information
If you have any questions or comments about this advisory:
- Open an issue in this repo
- DM @ndm on twitter
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 💎RubyGems | secure_headers | ≥ 6.0.0&&< 6.2.0 | 6.2.0bundle update secure_headers --conservative |
| 💎RubyGems | secure_headers | ≥ 5.0.0&&< 5.1.0 | 5.1.0bundle update secure_headers --conservative |
| 💎RubyGems | secure_headers | all versions | 3.8.0bundle update secure_headers --conservative |
Affected Products
secure headerstwitterResearch use only. For defensive security, authorized penetration testing, and academic research only. Never execute exploit code against systems without explicit written authorization.
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for secure_headers, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update secure_headers to 6.2.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2020-5217 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
Fixing This On Your OS
If you run this on a Linux distribution, patch through your package manager against the distro's own security advisory below — it tracks the exact backported fix for your release, which can ship on a different timeline (and sometimes a different severity) than the upstream project.
Satellite 6 ships Secure Header rubygem, however it does not accept any user input in override_content_security_policy_directive or append_content_security_policy_directive. All directives are hard-coded and therefore, Satellite 6 is not vulnerable to this CVE. We may update this rubygem in future release.
| Product | Fixed in | Advisory |
|---|---|---|
| Red Hat Satellite 6.8 for RHEL 7 | ansible-collection-redhat-satellite-0:1.3.0-1.el7sat | RHSA-2020:4366 |
Frequently Asked Questions
Is CVE-2020-5217 in your dependencies?
Find it across RubyGems, including transitive dependencies.