Skip to content

virtualenv activation scripts run code from a path that contains a placeholder name

High severity GitHub Reviewed Published Oct 1, 2026 in pypa/virtualenv • Updated Oct 21, 2026

Package

pip virtualenv (pip)

Affected versions

>= 20.26.6, <= 21.14.1

Patched versions

21.14.2

Description

Impact

ViaTemplateActivator.instantiate_template replaces the placeholders in an activation template one key at a time, each across the whole text. When a value that it already substituted and quoted contains the name of a later placeholder, the later pass rewrites that name inside the quoted literal. The inserted value brings its own quotes, which close the outer literal, and the rest of the path runs as code when a user activates the environment.

__VIRTUAL_NAME__ holds the environment's directory name, so a destination such as A__VIRTUAL_NAME__B/x'$(touch PWNED)'y is enough. In the default activate for bash, the VIRTUAL_ENV line renders as:

VIRTUAL_ENV='/tmp/t/A'x'"'"'$(touch PWNED)'"'"'y'B/x'"'"'$(touch PWNED)'"'"'y'

Sourcing that script runs touch PWNED. The same payload in a path without the placeholder name runs nothing. activate.fish runs the command the same way, and the reporter also showed Python code execution through activate.xsh and bin/activate_this.py. In activate.csh, a directory named after a prompt display placeholder, such as __VIRTUAL_PROMPT_DISPLAY_TCSH__, breaks the script with Unmatched '''.

No change to quote() can stop this, because the rewrite happens after quote() ran; it defeats shlex.quote and repr() alike.

An attacker needs to influence the destination path or --prompt: a CI job or wrapper tool that builds the path from a branch or project name, or a directory in a cloned repository or unpacked archive.

The bug affects versions 20.26.6 through 21.14.1. Version 20.26.6 first quoted template values (#2768) and introduced the per-key loop.

Patches

Fixed in 21.14.2. Activation scripts now keep a placeholder name in the environment's path or prompt as plain text, and activating the example above runs nothing. Fix: pypa/virtualenv@602b266

Workarounds

Do not create an environment whose path or --prompt contains a placeholder name such as __VIRTUAL_NAME__, __VIRTUAL_ENV__ or __VIRTUAL_PROMPT__, and inspect the generated activation script before you source one you did not create.

Credits

Kevin Stubbings (@Kwstubbs) of GitHub Security Lab reported this issue.

References

@gaborbernat gaborbernat published to pypa/virtualenv Oct 1, 2026
Published by the National Vulnerability Database Oct 20, 2026
Published to the GitHub Advisory Database Oct 21, 2026
Reviewed Oct 21, 2026
Last updated Oct 21, 2026

Severity

High

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v3 base metrics

Attack vector
Local
Attack complexity
Low
Privileges required
None
User interaction
Required
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
High

CVSS v3 base metrics

Attack vector: More severe the more the remote (logically and physically) an attacker can be in order to exploit the vulnerability.
Attack complexity: More severe for the least complex attacks.
Privileges required: More severe if no privileges are required.
User interaction: More severe when no user interaction is required.
Scope: More severe when a scope change occurs, e.g. one vulnerable component impacts resources in components beyond its security scope.
Confidentiality: More severe when loss of data confidentiality is highest, measuring the level of data access available to an unauthorized user.
Integrity: More severe when loss of data integrity is the highest, measuring the consequence of data modification possible by an unauthorized user.
Availability: More severe when the loss of impacted component availability is highest.
CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

EPSS score

Exploit Prediction Scoring System (EPSS)

This score estimates the probability of this vulnerability being exploited within the next 30 days. Data provided by FIRST.
(15th percentile)

Weaknesses

Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')

The product constructs all or part of an OS command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended OS command when it is sent to a downstream component. Learn more on MITRE.

Improper Control of Generation of Code ('Code Injection')

The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment. Learn more on MITRE.

CVE ID

CVE-2026-113776

GHSA ID

GHSA-8rjx-v5ww-45pp

Source code

Credits

Loading Checking history
See something to contribute? Suggest improvements for this vulnerability.