Skip to content

`datamodel-code-generator` vulnerable to code injection via unescaped carriage return in GraphQL Union description

High severity GitHub Reviewed Published Jun 8, 2026 in datamodel-code-generator/datamodel-code-generator • Updated Jul 28, 2026

Package

pip datamodel-code-generator (pip)

Affected versions

>= 0.25.0, < 0.60.1

Patched versions

0.60.1

Description

Summary

datamodel-code-generator is vulnerable to code injection when generating Python models from an attacker-controlled GraphQL schema. A description on a Union type, written in the regular-string form ("...") with a literal \r escape, is rendered into a Python # comment by a Jinja2 filter that handles only \n. Python's tokenizer treats a bare CR as a physical-line terminator, so the comment ends at the \r and the text after it is parsed as module-level Python. The injected code executes at import time of the generated .py, in the context of any consumer that imports the model. No special CLI flags are required.

This affects versions >=0.25.0, <0.60.1 and is fixed in 0.60.1.

Details

The vulnerable output was generated by the GraphQL Union templates:

  • src/datamodel_code_generator/model/template/UnionTypeStatement.jinja2
  • src/datamodel_code_generator/model/template/UnionTypeAliasAnnotation.jinja2
  • src/datamodel_code_generator/model/template/UnionTypeAliasType.jinja2

Before 0.60.1, Union descriptions were rendered as Python comments using logic equivalent to # {{ description | replace('\n', '\n# ') }}.

The Jinja2 replace filter handled \n, but bare \r was not normalized. Python's lexical analysis treats CR, LF, and CRLF as physical line terminators. As a result, a malicious GraphQL schema could cause the generated Python comment to end early and emit attacker-controlled text as module-level Python code.

The description value reaches Union generation from graphql-core through description=union_object.description. GraphQL regular string values can contain \r escapes, and graphql-core preserves that value. Block strings are not affected in the same way because their line endings are normalized before reaching code generation.

Impact

An attacker who can provide or influence a GraphQL schema processed by datamodel-code-generator could cause arbitrary Python code to be generated into the output file. That code would execute with the privileges of the process importing the generated model.

This can affect developers, CI pipelines, or applications that run datamodel-codegen --input-file-type graphql on untrusted or third-party GraphQL schemas and then import the generated Python module.

No custom templates, special CLI flags, or remote reference options are required.

Remediation

Upgrade to datamodel-code-generator 0.60.1 or later.

The fix normalizes carriage returns in GraphQL Union descriptions before rendering them as Python comments, so injected text remains inside the comment block.

Resolution

The fix applies comment_safe to GraphQL Union descriptions before template rendering. It normalizes CRLF and bare CR to LF so the existing Union comment templates keep the whole description inside the generated Python comment block.

Submitted by: Hamza Haroon (thegr1ffyn)

References

Published to the GitHub Advisory Database Jul 28, 2026
Reviewed Jul 28, 2026
Last updated Jul 28, 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.
(10th percentile)

Weaknesses

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.

Improper Neutralization of Special Elements Used in a Template Engine

The product uses a template engine to insert or process externally-influenced input, but it does not neutralize or incorrectly neutralizes special elements or syntax that can be interpreted as template expressions or other code directives when processed by the engine. Learn more on MITRE.

CVE ID

CVE-2026-54621

GHSA ID

GHSA-j884-q54q-mmx3

Credits

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