EUVD-2025-18458
CVSS Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Lifecycle Timeline
5Tags
Description
Conda-build contains commands and tools to build conda packages. Prior to version 25.3.0, the pyproject.toml lists conda-index as a Python dependency. This package is not published in PyPI. An attacker could claim this namespace and upload arbitrary (malicious) code to the package, and then exploit pip install commands by injecting the malicious dependency in the solve. This issue has been fixed in version 25.3.0. A workaround involves using --no-deps for pip install-ing the project from the repository.
Analysis
Conda-build versions prior to 25.3.0 are vulnerable to dependency confusion/namespace squatting attacks where an attacker can claim the unpublished 'conda-index' package on PyPI and inject malicious code that gets installed when users run pip install on conda-build projects. This is a critical supply-chain attack vector with CVSS 9.8 (CRITICAL) affecting all users who install conda-build from source or install projects that depend on it via pip, potentially compromising developer environments and CI/CD pipelines. The vulnerability is network-accessible, requires no privileges or user interaction, and provides complete system compromise (confidentiality, integrity, availability).
Technical Context
The vulnerability stems from CWE-1357 (Weak Dependency Resolution), a supply-chain attack class where packages declare dependencies that don't exist in the primary package repository. Conda-build's pyproject.toml explicitly lists 'conda-index' as a required Python dependency, but this package was never published to PyPI. When pip resolves dependencies during installation, it searches PyPI first; an attacker can register 'conda-index' on PyPI and upload arbitrary malicious code. The root cause is broken dependency management—the conda-build project depends on 'conda-index' which appears to be an internal/private package that should either be: (1) properly published and maintained, (2) vendored into conda-build, or (3) made optional. This affects conda-build CPE: cpe:2.3:a:conda:conda-build:*:*:*:*:*:*:*:* for versions < 25.3.0. The attack exploits the implicit trust users place in dependency resolution and the namespace sprawl of PyPI.
Affected Products
Product: Conda-build, Vendor: Conda (Anaconda Inc.), Affected Versions: All versions prior to 25.3.0 (CPE: cpe:2.3:a:conda:conda-build:*:*:*:*:*:*:*:* where version < 25.3.0). Specific known affected versions include at least 25.2.0 and earlier releases. Secondary impact affects any project with a pyproject.toml or setup.py that declares a dependency on conda-build < 25.3.0 (transitive dependency chain). Users installing conda-build via: (1) pip install conda-build (< 25.3.0), (2) pip install -e . in conda-build source directory (< 25.3.0), or (3) any tool that auto-installs conda-build as a build backend are at risk. The vulnerability does NOT affect users who install conda-build via conda package manager (conda install conda-build) as that uses a different resolver.
Remediation
Immediate actions: (1) Upgrade to conda-build >= 25.3.0 immediately (patch version 25.3.0 fixes the issue by removing or properly declaring the conda-index dependency). Run: pip install --upgrade conda-build>=25.3.0. (2) For users unable to patch immediately, apply the workaround: pip install --no-deps conda-build (installs without resolving transitive dependencies; requires manual installation of actual required dependencies). (3) Verify your environment: check installed package versions with pip show conda-build; if version < 25.3.0, prioritize upgrade. (4) For CI/CD pipelines: update dependency specifications (requirements.txt, pyproject.toml, setup.cfg) to enforce conda-build>=25.3.0 or use conda package manager instead of pip for conda-build installation. (5) Audit PyPI for any malicious 'conda-index' packages uploaded during the vulnerability window and report to PyPI abuse team. (6) Check pip/pip freeze logs and auditing tools for unexpected 'conda-index' package installations in your environment history.
Priority Score
Share
External POC / Exploit Code
Leaving vuln.today