When a request returns a 500, jupyter_server/log.py logs a small JSON block of request headers.
The Referer header was copied into it as-is, so a token in the Referer URL ended up in the logs in plain text.
Anyone who can read the server logs can pick tokens out of these 500 entries. Tokens end up in the Referer during normal token-based login and launch flows.
Affected: all versions before 2.21.0.
Any malformed request that returns a 500 works:
curl -i -X POST \
-H 'Content-Type: application/json' \
-H 'Referer: http://127.0.0.1:8899/tree?token=REFERTOKEN' \
--data '{"name":123}' \
'http://127.0.0.1:8899/api/kernels?token=VALIDTOKEN'
The log shows the token twice, once raw and once redacted:
"Referer": "http://127.0.0.1:8899/tree?token=REFERTOKEN",
[E ... ServerApp] 500 POST /api/kernels?token=[secret] (...) referer=http://127.0.0.1:8899/tree?token=[secret]
Fixed in 2.21.0 by 5251352. Header values are now scrubbed before the block is logged. Upgrade to 2.21.0 or later.
Limit who can read the server logs. If you can, avoid flows that put the token in the URL.
{
"cwe_ids": [
"CWE-532"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-17T20:28:55Z",
"nvd_published_at": null,
"severity": "HIGH"
}