Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessGain control of a package index
-
DeliveryServe crafted index data with forged Saved line
-
ExploitUnverified archive written to wheel cache
-
ExecutionMismatched archive selected without digest check
-
ImpactInstall unverified wheel into environment
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A08:2025 - Software or Data Integrity Failures
ASVS 5.0
V15 - Secure Coding and Architecture Integrity verification
|
| Exploitation Conditions | Exploitation requires an attacker-controlled package index and affects only virtualenv 21.14.8 during its wheel-download/seeding path. The forged-line variant additionally requires default-index digest comparison to be enabled, the ability to inject a misleading pip 'Saved' line through a discarded link's data-requires-python attribute, AND knowledge of the absolute filesystem path to the genuine bundled wheel. The separate rejected-cache-file variant instead requires control over the downloaded data or archive selection but not the forged-line prerequisites. Significant limiting factors: the issue explicitly does not bypass TLS and does not grant control of default PyPI (so the attacker must already be positioned to serve a malicious index), explicit custom-index exceptions are outside the promised PyPI comparison, and successful attack depends on the victim running the vulnerable virtualenv seeding (UI:P) with high overall complexity (AC:H/AT:P). |
| Impact | This is a genuinely low-priority issue whose signals are consistent rather than conflicting. The vendor CVSS 4.0 score of 2.3 (AV:N/AC:H/AT:P/PR:N/UI:P, impacts VC:L/VI:L/VA:L) reflects a hard-to-reach integrity-check weakness (CWE-354) rather than a readily exploitable flaw. There is no CISA KEV listing and no public exploit identified at time of analysis - only the upstream fix commit and GHSA advisory are available. EPSS was not supplied in the input. No SSVC decision record was provided. The high attack complexity (AC:H), the presence of a specific attack requirement (AT:P), and the dependency on passive user interaction (UI:P) all point to a narrow, precondition-heavy attack, not an automatable or widespread one. Crucially, the description itself states the issue 'does not independently bypass TLS or establish control of default PyPI,' so an attacker must already be positioned to control a package index - a substantial prerequisite that is not captured in a single raw metric. Net: real but marginal risk, appropriately rated low; not a patch-tonight item absent a threat model involving untrusted or compromised indexes. |
| Exploit ScenarioDerived | An attacker who already controls a package index that a developer's virtualenv seeding talks to serves a crafted index response: for the forged-line variant they inject a misleading pip 'Saved' line via a discarded link's data-requires-python attribute (also needing the absolute path to the genuine bundled wheel), or for the simpler variant they manipulate the downloaded data/archive selection. Because download_wheel() writes downloads into the reusable wheel cache before verifying, a digest-mismatched archive is left in the cache and can later be selected and installed without verification. No public exploit code is known, and the chain requires high complexity plus the victim running an affected virtualenv (21.14.8). |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 20, 2026
Wheel-cache poisoning in virtualenv 21.14.8 can leave a rejected, attacker-supplied archive in the reusable wheel cache, where a later seeding operation may select it without verification. The forged pip 'Saved' line path requires an attacker-controlled package index, default-index digest comparison, and knowledge of the absolute path to the genuine bundled wheel; the separate rejected-cache-file variant requires control over downloaded data or archive selection. The issue is low severity, does not bypass TLS or establish control of default PyPI, and is fixed in 21.14.9; no public exploit code is identified in the available data.
RemediationAI
Upgrade to virtualenv 21.14.9, which stages downloads in a private temporary directory, verifies the staged archive, and only then adds it to the reusable wheel cache; see the 21.14.9 release (https://github.com/pypa/virtualenv/releases/tag/21.14.9), advisory GHSA-494p-63j8-28p3, and fixing commit. If upgrading is not immediately possible, avoid running vulnerable 21.14.8 seeding against attacker-influenced or custom package indexes, restrict PIP_INDEX_URL and related configuration to trusted default PyPI, and rebuild or clear the virtualenv app-data wheel cache so previously rejected archives cannot be selected later. Those workarounds reduce exposure but do not address the rejected-cache-file variant if downloaded data or archive selection is controlled, and clearing the cache forces fresh downloads and slows subsequent seeding.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in pypi
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
virtualenv's wheel seeder invokes pip download to fetch seed wheels and, in 21.14.8, download_wheel() wrote package-index downloads into the reusable wheel cache before determining the verification target from pip's textual Saved output. A controlled package index can inject a misleading Saved line through a discarded link's data-requires-python attribute, causing the wrong path to be hashed; a digest mismatch then leaves the rejected archive in cache for later selection without verification. This is CWE-354 (Improper Validation of Integrity Check Value): the code validates a digest against a path selected from untrusted textual output rather than against the archive it intends to install. The fix in 21.14.9 stages downloads in a private temporary directory, selects the archive from that directory instead of pip output, and only moves verified files into the cache; the pip seeder uses the same path. Explicit custom-index exceptions are outside the promised PyPI comparison. The advisory tags the issue as Authentication Bypass and Python, but the assessed CVSS 3.1 vector is integrity-only (C:N/I:L/A:N) and does not grant default-PyPI control or TLS bypass.
Know when Python gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:P/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N
Remote but requires attacker-controlled index, absolute-path knowledge and victim seeding, so AC:H and UI:R; PR:N as no victim-side privilege is needed; impact is a low integrity hit from an unverified wheel.
Primary rating from Vendor (github).
CVSS VectorVendor: github
DescriptionCVE.org
virtualenv is a tool for creating isolated virtual python environments. From 21.14.8 until 21.14.9, download_wheel() writes package-index downloads directly into the reusable wheel/house cache before selecting a verification target from pip's textual Saved output. With default-index digest comparison enabled, a controlled package index can inject a misleading Saved line through a discarded link's data-requires-python attribute; this forged-line variant also requires knowledge of an absolute path to the genuine bundled wheel. The wrong path can then be hashed, while a digest mismatch leaves the rejected archive in the cache for later selection without verification. A separate rejected-cache-file variant requires control over downloaded data or archive selection but not the forged-line prerequisites. Explicit custom-index exceptions are outside the promised PyPI comparison, and the issue does not independently bypass TLS or establish control of default PyPI. This issue is fixed in version 21.14.9.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
EUVD-2026-105260
GHSA-494p-63j8-28p3
NVD
CVE.org
EPSS
VulDB
Exploit-DB