Supply chain · pip · High

Critical Malicious Package 'requests-crypt' in PyPI Contains Hidden Backdoor

Threat Engine Auto-Feed · data current as of 2026-08-21

A malicious package, 'requests-crypt' version 0.1.0, has been identified in the PyPI ecosystem. It contains a hidden backdoor that executes arbitrary code based on HTTP response fields, posing a critical supply-chain risk for Python projects. Organizations using this package are exposed to potential remote code execution.

  • Package: pip / requests-crypt
  • Affected versions: = 0.1.0
  • Severity: critical
  • Reach: unknown
  • Exposure: 65/100 (High)

What happened — Credential / secret theft

The 'requests-crypt' package (version 0.1.0) in the PyPI ecosystem was found to contain malicious code. While it purports to be an HTTP request library with additional functions, it includes a hidden backdoor. This backdoor secretly checks for specific fields in HTTP responses and, if found, executes their content.

How the attack works

This is a supply-chain attack where a seemingly legitimate software package is compromised to deliver malicious functionality. The attack leverages the trust placed in open-source packages. When the malicious 'requests-crypt' package is installed and used, its hidden code monitors network traffic. Upon detecting specific conditions in HTTP responses, it triggers the execution of content from those responses, allowing an attacker to achieve remote code execution without direct interaction with the compromised system.

Who's exposed

Profiles most at risk

  • Development teams using Python and PyPI for their projects
  • Organizations with CI/CD pipelines that automatically pull dependencies, especially if unpinned
  • Any system or application that has 'requests-crypt' version 0.1.0 installed as a dependency

Conditions that increase exposure

  • Unpinned dependencies in `requirements.txt` or `pyproject.toml` that could resolve to the malicious version
  • Lack of dependency lockfiles (`poetry.lock`, `Pipfile.lock`) or not committing them to version control
  • Use of internal package mirrors that do not filter known malicious packages
  • Automated build processes that install dependencies without prior security scanning

Blast-radius scenarios

  • An attacker could achieve remote code execution on systems where the package is installed and used, potentially leading to data exfiltration, system compromise, or further network penetration.
  • Compromised build environments or developer workstations if the package is installed there.
  • The blast radius extends to any production, staging, or development environment where this specific malicious version of 'requests-crypt' is present and actively used.

What to do (defensive)

Detect

  • Scan dependency trees for 'requests-crypt' version 0.1.0 using software composition analysis (SCA) tools.
  • Review `requirements.txt`, `pyproject.toml`, `Pipfile`, and lockfiles for direct or transitive dependencies on 'requests-crypt' version 0.1.0.
  • Monitor network traffic for unusual outbound connections or command and control (C2) activity from applications using this package.

Contain

  • Immediately quarantine or isolate any systems identified as having 'requests-crypt' version 0.1.0 installed.
  • Block network access for affected systems to prevent potential data exfiltration or further compromise.
  • Prevent new installations of 'requests-crypt' version 0.1.0 by configuring package managers or internal registries to deny it.

Remediate

  • Remove 'requests-crypt' version 0.1.0 from all affected environments and applications.
  • Replace the package with a known safe alternative if its functionality is critical, or remove the dependency entirely if not needed.
  • Rotate any credentials or secrets that might have been exposed on compromised systems.
  • Rebuild affected applications and redeploy them after ensuring all malicious components have been removed and dependencies are secure.

Frontier verdict — High

Prioritize immediate detection and removal of 'requests-crypt' version 0.1.0 due to its critical backdoor functionality and potential for remote code execution.

Sources

For detection-engineering and awareness only · point-in-time · not security advice · sourced from the GitHub Advisory Database.