# CVE-2026-104073: NetBox server-side template injection via custom links

Source: https://junglewise.ai/threats/cve-2026-104073-netbox-versions-2-9-5-before-4-7-0-contain-a-server-side  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-06. Updated: 2026-10-06  
License: CC BY 4.0 with a link back

CVE-2026-104073 · Severity: high · CVSS 7.6 · Published 2026-10-06

NetBox is an open-source inventory and documentation platform used by IT teams to manage network infrastructure. A server-side template injection vulnerability in versions before 4.7.0 allows low-privileged users with "Can add custom links" permission to steal session cookies and API tokens from other users, leading to full account takeover when a privileged user views a crafted malicious link.

Vendors: [NetBox Community](https://junglewise.ai/threats/vendors/netbox-community).

## Quick answers

**Is CVE-2026-104073 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-104073 being exploited in the wild.

**How severe is CVE-2026-104073?** High severity, with a CVSS score of 7.6, published 6 October 2026.

**Which products and versions are affected?** NetBox Community NetBox 2.9.5 before 4.7.0.

**Is there a fix for CVE-2026-104073?** Yes: fixed in 4.7.0. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-104073 attacked?** It is attacked over the network, in a web application.

**Am I affected by CVE-2026-104073?** Check whether you run NetBox Community NetBox at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- Fixed in 4.7.0

## Technical details

The vulnerability exists in NetBox's custom link rendering, which passes the raw Django HttpRequest object to a Jinja2 template context without sanitization. An attacker with "Can add custom links" permission can embed sensitive request attributes (such as request.COOKIES['sessionid'] or API tokens) into template expressions that bypass the clean_html sanitizer and auto-exfiltrate them via image src URLs when a privileged user accesses the object. A fix was implemented in version 4.7.0 by introducing the get_safe_request_context() utility function that whitelists only safe request attributes.

## Affected products

- NetBox Community NetBox 2.9.5 before 4.7.0

## Timeline

- 2026-07-03: disclosed
- 2026-07-07: patched

## References

- [https://github.com/netbox-community/netbox/issues/22607](https://github.com/netbox-community/netbox/issues/22607)
- [https://github.com/netbox-community/netbox/pull/22616](https://github.com/netbox-community/netbox/pull/22616)

## Related threats

- [NetBox broken access control in bulk component creation](https://junglewise.ai/threats/cve-2026-116845-netbox-from-2-9-0-before-4-7-4-contains-a-broken-access-control)