/

CVE-2026-75486

Severity:

HIGH RISK

(8.0)

OS Command Injection (CWE-78) in Synk Sweater Comb

CVE-2026-75486

Deferred

Vulnerability type

Category:

OS Command Injection

CWE:

CWE-78

Impact:

Arbitrary Command Execution

Severity:

HIGH RISK

(8.0)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Synk Sweater Comb contains a command injection flaw in the handling of a Git branch name from the .vervet.yaml configuration file. If an attacker can control that configuration and cause the lint command to run, arbitrary operating system commands may execute with the privileges of the process.

Affected context

The vulnerability occurs in repositories or build environments where Synk Sweater Comb processes a .vervet.yaml file containing the linters.<key>.optic-ci.original branch name. Risk depends on an attacker being able to modify that configuration and on the affected lint workflow executing the branch-checking logic.

Exploit preconditions

  • Attack surface: The lint command and its branch validation functionality
  • Required access: Ability to control or modify the relevant .vervet.yaml configuration
  • Trigger point: Execution of the lint workflow that invokes expectGitBranch()
  • Why exploitation is possible: The branch value is inserted into a shell command without adequate escaping or command-argument separation

Why this happens

The underlying bug is OS command injection caused by combining attacker-influenced configuration data with a shell command through an unescaped template literal. The expectGitBranch() function passes the resulting string to child_process.exec(), allowing shell metacharacters in the branch value to alter command interpretation. This behavior maps to CWE-78: Improper Neutralization of Special Elements used in an OS Command.

Potential impact

  • An attacker may execute arbitrary operating system commands
  • The local development or CI environment could be compromised
  • Source code, credentials, or build secrets could be accessed
  • Build artifacts or repository contents could be modified
  • Malicious changes could propagate through development or supply-chain workflows

General mitigation guidance

Mitigation generally focuses on preventing configuration data from being interpreted as shell syntax.

Common defensive patterns include:

  • Using APIs that pass command arguments as structured values rather than constructing shell command strings
  • Avoiding shell invocation when a direct process or library API is available
  • Treating repository configuration as untrusted input unless its integrity is established
  • Validating branch names against the expected Git reference format
  • Running linting and CI tasks with least-privilege accounts and isolated credentials
  • Restricting access to configuration files and protecting build secrets from the linting environment

How teams detect issues like this

Organizations can look for this vulnerability class through:

  • Static analysis rules identifying child_process.exec() calls that incorporate configuration or repository data
  • Code review checks for template literals or string concatenation used to construct shell commands
  • Searches for untrusted values passed to shell-enabled process APIs
  • Tests using metacharacter-containing inputs to verify that values remain data rather than commands
  • CI policies that flag shell execution with unsanitized configuration-derived arguments
  • Automated security review tools that recognize command construction and tainted data flows

Sources

SourceLink
github.comgithub.com/snyk/sweater-comb/commit/05a0eec4f2acb9ce6d4814016b475504fc64eab2
github.comgithub.com/snyk/sweater-comb/pull/743
github.comgithub.com/snyk/sweater-comb/releases/tag/v3.8.8
vulncheck.comvulncheck.com/advisories/synk-sweater-comb-command-injection-via-vervet-yaml-branch-name

Frequently asked

questions

It is not necessarily exposed as a direct network service. Exploitation generally depends on an attacker controlling the repository configuration and causing the affected lint command to run, such as in a development or CI environment.

The attacker needs the ability to modify or supply the relevant .vervet.yaml content. The practical impact also depends on the privileges, network access, and secrets available to the process running the lint command.

Avoid shell command construction for attacker-influenced values. Prefer structured process APIs or Git library functions, and validate branch names against a narrowly defined expected format.

Treat configuration and repository metadata as untrusted, review all shell execution paths, enforce least privilege in CI, and use automated analysis to detect tainted values reaching command-execution APIs.

Find out what's already
exploitable in your codebase.