Skip to main content
← Vulnerabilities

labbench before 0.46.1 Deserialization of Untrusted Data via read() Pickle Auto-Dispatch

labbench before 0.46.1 is affected when read() loads files with the 'p' or 'pickle' extension, which pass to pandas.read_pickle. Fixed in 0.46.1.

Written by anthropic/claude-haiku-5.5 from the sources below. Verify against them.

8.5HighCNA
Exploited
No known exploitation
Fix
Fixed in 0.46.1
Who is exposed
Specific setups only

Do first: Check whether your setup matches what the bug needs. labbench 0.46.0 source: read() maps the 'p' and 'pickle' extensions directly to pandas.read_pickle

14Pcve.now Index 14 of 100 · methodology v3 · provisional
  • Exploitation?/40Exploitation: unknown (data gap). No CISA SSVC exploitation rating yet
  • Exposure5/25Specific setups only
  • Reach?/20Reach: unknown (data gap). No tracked technology or package matched
  • Severity8.5/10CVSS 8.5
  • No patch0/5No signal

What would change the score

  • Would score about 64 if added to CISA KEV
  • Would score about 59 if CISA's SSVC rates exploitation active
  • Would score about 24 if CISA's SSVC rates it automatable

CVE-2026-120833 is not in CISA's Known Exploited Vulnerabilities (KEV) catalog. No public exploit code is tracked. It is rated CVSS 8.5 (high) and affects labbench; fixed in 0.46.1.

Published
Last modified
Published
Not listed in CISA KEV
0.14% · 3rd percentile
14/100 · provisional
Specific setups only
Fix
Fixed in 0.46.1
None on Exploit-DB, Metasploit or GitHub

Summary

The labbench Python package (versions before the 0.46.1 fix) is affected. Its read() and read_relational() functions automatically treat any file ending in .p or .pickle as a pickle file, so a maliciously crafted data file or archive can run arbitrary code on the user's computer when it is loaded.

Read the original description

labbench before 0.46.1 contains a deserialization of untrusted data vulnerability in read() and read_relational() that passes any file with a .p or .pickle extension to pandas.read_pickle based on the extension alone. Attackers who supply a crafted data file or relational data archive can execute arbitrary code when a user loads it with these functions.

What to do

  • Check whether your setup matches what the bug needs.

    labbench 0.46.0 source: read() maps the 'p' and 'pickle' extensions directly to pandas.read_pickle
  • Upgrade to a fixed version: 0.46.1.

Timeline

  1. CVE published
  2. First seen by cve.now

Affected versions

  • NIST labbench

    fixed in 0.46.1

Evidence 7

  1. Patch
  2. Patch
  3. Repository
  4. Repository
  5. Repository
Show all 7 sourcesShow fewer sources
  1. Link
  2. Link

Corrections

None so far.

Something wrong on this page? Report an error. Every fix is listed here.

FAQ

Questions about CVE-2026-120833

Is CVE-2026-120833 being exploited?

It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data

What does this mean?

Loading a crafted pickle file can execute arbitrary code, because unpickling can execute arbitrary code. In labbench 0.46.0, read() sends files with the 'p' or 'pickle' extension directly to pandas.read_pickle.Source: github.com

Which versions are affected by CVE-2026-120833?

labbench: before 0.46.1.Source: cve.now feed data

Am I affected, and how do I check?

labbench versions before 0.46.1 are affected. To check, look at the installed labbench version and whether code calls read() on files ending in .p or .pickle.Source: github.com

How do I fix CVE-2026-120833?

Upgrade to labbench 0.46.1, which is the fixed version.Source: github.com