CVE-2026-54653: Datamodel code generator `datamodel-code-generator` Vulnerable To Code Injection Vulnerability
HIGHCVE-2026-54653 is a HIGH (CVSS 8.8) `datamodel-code-generator` vulnerable to code injection vulnerability (CWE-94: Improper Control of Generation of Code ('Code Injection')) affecting Datamodel code generator libraries in the PyPI ecosystem. The issue involves `datamodel-code-generator` vulnerable to code injection in via attacker-controlled `default_factory` schema field. Remediation is available by upgrading affected packages to version 0.60.2.
CVE-2026-54653 at a Glance
Security Impact
CVSS: 8.8
The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H).
How to Fix CVE-2026-54653
To remediate this vulnerability, upgrade all affected Datamodel code generator dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.
pip install datamodel-code-generator>=0.60.2
Run the following command in your project directory to verify that all resolved dependencies use the patched release:
pip show datamodel-code-generator
How to Check if You Are Affected
To determine whether your Python environment is affected by CVE-2026-54653, inspect your installed packages using pip:
pip show datamodel-code-generator
Analyze the terminal output to locate any version of Datamodel code generator in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 0.60.2), your application is affected.
Note: The pip show datamodel-code-generator command checks currently installed packages in your active Python environment.
Technical Details
Vulnerability Mechanism & Technical Analysis
This vulnerability is classified under CWE-94: Improper Control of Generation of Code ('Code Injection'), where 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. In Datamodel code generator, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.
Source Advisory Statement
CVE-2026-54653 Affected Datamodel code generator Versions
1 component(s)| Product / Package | Ecosystem | Affected Versions | Fixed Version |
|---|---|---|---|
|
datamodel-code-generator
|
PyPI |
≥ 0.17.0
< 0.60.2
|
0.60.2
|
References
Frequently Asked Questions
What is CVE-2026-54653?
Which Datamodel code generator versions are affected?
What version fixes CVE-2026-54653?
Is datamodel-code-generator affected?
What is the CVSS score of CVE-2026-54653?
What CWE is associated with CVE-2026-54653?
How can I check PyPI dependencies for CVE-2026-54653?
Machine-readable OSV JSON
OSV v1.5 Schema
{
"schema_version" : "1.4.0",
"id" : "GHSA-386q-5hp3-95m9",
"modified" : "2026-07-28T21:48:14Z",
"published" : "2026-07-28T21:48:14Z",
"aliases" : [ "CVE-2026-54653" ],
"summary" : "`datamodel-code-generator` vulnerable to code injection in via attacker-controlled `default_factory` schema field",
"details" : "### Summary\n\n`datamodel-code-generator` is vulnerable to code injection when generating Python models from an attacker-controlled JSON Schema, OpenAPI, YAML, JSON, Avro, Protobuf, or XSD schema. When a property carries a `\"default_factory\"` key, its value is interpolated verbatim — as a raw Python expression — into the generated `Field(default_factory=...)` / `field(default_factory=...)` call. Because this assignment is evaluated at class-definition time (i.e. on `import` of the generated module), an attacker who controls the schema controls a Python expression that runs in the consumer's process. No special CLI flags are required.\n\n### Details\n\nThe vulnerable chain spans the JSON-Schema-shaped parser and three sink locations (Pydantic v2, dataclass, msgspec):\n\n**Source — schema → `extras`**:\n\n- `src/datamodel_code_generator/parser/jsonschema.py:600-614` — `DEFAULT_FIELD_KEYS` includes the literal string `\"default_factory\"`.\n- `src/datamodel_code_generator/parser/jsonschema.py:457-459` — `JsonSchemaObject.__init__` stores any non-standard key (including `default_factory`) in `self.extras`.\n- `src/datamodel_code_generator/parser/jsonschema.py:797-812` — `get_field_extras` preserves `default_factory` through to the field model.\n\n**Sinks — `extras` → generated Python expression**:\n\n1. `src/datamodel_code_generator/model/pydantic_base.py:222-249`:\n\n ```python\n default_factory = data.pop(\"default_factory\", None)\n ...\n if default_factory is not None:\n field_arguments = [f\"default_factory={default_factory}\", *field_arguments]\n ```\n\n The `default_factory` value is interpolated raw (no `repr()`, no validation).\n\n2. `src/datamodel_code_generator/model/dataclass.py:211`:\n\n ```python\n f\"{k}={v if k == 'default_factory' else repr(v)}\"\n ```\n\n Explicit special-case to skip `repr()` for `default_factory`.\n\n3. `src/datamodel_code_generator/model/msgspec.py:361` — same pattern as dataclass.\n\nBecause `default_factory` is in `DEFAULT_FIELD_KEYS`, no special CLI flag is needed to reach the sink. Any input format that uses the JSON-Schema-shaped parser (`jsonschema`, `openapi`, `yaml`, `json`, `dict`, `csv`) — and any input format that converts to it (`avro`, `protobuf`, `xmlschema`) — is in scope.\n\n### Confirmed PoC matrix\n\n| Input file type | Output model type | Result |\n|---|---|---|\n| `jsonschema` | `pydantic_v2.BaseModel` | RCE on import |\n| `jsonschema` | `dataclasses.dataclass` | RCE on import |\n| `jsonschema` | `msgspec.Struct` | RCE on import |\n| `jsonschema` | `typing.TypedDict` | safe (TypedDict doesn't render `field()`; `default_factory` silently dropped) |\n| `openapi` | `pydantic_v2.BaseModel` | RCE on import |\n\nOther JSON-Schema-shaped inputs (`yaml`, `json`, `dict`, `csv`, `avro`, `protobuf`, `xmlschema`) follow the same code path and are expected to reproduce.\n\n### PoC\nSelf contained Proof of Concept is available at my secret gist: https://gist.github.com/thegr1ffyn/9648b0fe4fcf7d569ac8e61dd11eebaf\n\n### Impact\n\n- **Who's affected**: any developer or CI pipeline that runs `datamodel-codegen` against a schema they didn't author themselves — third-party API specs, schemas pulled from a registry, vendored upstream `.json` / `.yaml` / `.avsc` / `.proto` / `.xsd` files, schemas fetched from a remote URL or introspection endpoint — *and* who imports the generated `.py`.\n- **What it gains**: arbitrary Python code execution in the importer's process at `import` time. The PoC copies `/etc/passwd` to a tmp file to demonstrate arbitrary read; the same primitive supports any operation the importing process can perform (filesystem write, environment exfiltration, secondary network calls, RCE on CI runners).\n- **What it does NOT need**: no special CLI flags, no custom templates, no `--extra-template-data`, no `--use-schema-description`. Default invocation against a malicious schema is sufficient.\n- **What does block it**: choosing `--output-model-type typing.TypedDict` (which doesn't render `field()` / `Field()` calls). All other supported output model types are vulnerable.\n\n### Resolution\n\nThe fix validates schema-provided `default_factory` values while extracting JSON Schema field extras. Only the supported factory names `dict`, `list`, and `set` are accepted; any other value now raises a generator error before code generation. Generator-created default factories for supported mutable defaults and optional nested models continue to use the existing code paths.\n\n### Remediation\n\nUpgrade to `datamodel-code-generator` `0.60.2` or later.\n\nThis issue affects `datamodel-code-generator` versions `>= 0.17.0, <= 0.60.1` and is fixed in `0.60.2`.\n\nSubmitted by: Hamza Haroon (thegr1ffyn)",
"severity" : [ {
"type" : "CVSS_V3",
"score" : "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H"
} ],
"affected" : [ {
"package" : {
"ecosystem" : "PyPI",
"name" : "datamodel-code-generator"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0.17.0"
}, {
"fixed" : "0.60.2"
} ]
} ],
"database_specific" : {
"last_known_affected_version_range" : "<= 0.60.1"
}
} ],
"references" : [ {
"type" : "WEB",
"url" : "https://github.com/koxudaxi/datamodel-code-generator/security/advisories/GHSA-386q-5hp3-95m9"
}, {
"type" : "WEB",
"url" : "https://github.com/koxudaxi/datamodel-code-generator/commit/17fc235e234cbcfaaadef8c74cb72c9687db0d1d"
}, {
"type" : "PACKAGE",
"url" : "https://github.com/koxudaxi/datamodel-code-generator"
}, {
"type" : "WEB",
"url" : "https://github.com/koxudaxi/datamodel-code-generator/releases/tag/0.60.2"
} ],
"database_specific" : {
"cwe_ids" : [ "CWE-1336", "CWE-94" ],
"severity" : "HIGH",
"github_reviewed" : true,
"github_reviewed_at" : "2026-07-28T21:48:14Z",
"nvd_published_at" : null
}
}