10

CVSS Score

3.1

-

CVSS Score

Basic Information

Concerned about an active attack path?

Talk to our security experts and see Miggo in action.

Miggo Vulnerability Database

→

CVE-2025-70974

CVE-2025-70974: FASTJSON Includes Functionality from Untrusted Control Sphere

Fastjson before 1.2.48 mishandles autoType because, when an @type key is in a JSON document, and the value of that key is the name of a Java class, there may be calls to certain public methods of that class. Depending on the behavior of those methods, there may be JNDI injection with an attacker-supplied payload located elsewhere in that JSON document. This was exploited in the wild in 2023 through 2025. NOTE: this issue exists because of an incomplete fix for CVE-2017-18349. Also, a later bypass is covered by CVE-2022-25845.

(GitHub Advisory)

Miggo Vulnerability Database

→

CVE-2025-70974

CVE-2025-70974:

10

CVSS Score

3.1

-

CVSS Score

Basic Information

Is this CVE running in your environment?

Easily map the attack path and prioritize which CVEs are a threat to your organization

Validate Exposure

Technical Details

Package Name	Ecosystem	Vulnerable Versions	First Patched Version
com.alibaba:fastjson	maven	< 1.2.48	1.2.48

Technical Details

Vulnerability Intelligence
Miggo AI

Root Cause Analysis

The vulnerability lies in Fastjson's autoType feature, which allows specifying a Java class name in the JSON input using the @type key. The core of the issue is in the com.alibaba.fastjson.parser.ParserConfig.checkAutoType function. In vulnerable versions, this function did not properly validate the class name before attempting to load it. An attacker could provide the name of a malicious class (e.g., one that triggers a JNDI lookup) and Fastjson would load and instantiate it, leading to Remote Code Execution.

The patch in version 1.2.48 addresses this by introducing multiple layers of defense within checkAutoType. First, it inspects the class's bytecode using an ASM ClassReader to check for a @JSONType annotation before loading the class. This prevents the class's static initializers from running if it's not explicitly marked as safe for deserialization. Second, it tightens the conditions under which autoType is permitted, for example, by not trusting overly generic expected types like java.lang.Object or java.io.Serializable.

The analysis of the commits between versions 1.2.47 and 1.2.48 confirms these changes. The modifications in ParserConfig.java are central to the fix. The related changes in TypeUtils.loadClass (which performs the class loading) and MiscCodec.deserialze (a potential entry point for class loading) are supporting fixes to limit caching and reduce the attack surface. Therefore, any runtime profile during exploitation would prominently feature checkAutoType and its helper TypeUtils.loadClass.

Vulnerable functions

com.alibaba.fastjson.parser.ParserConfig.checkAutoType

src/main/java/com/alibaba/fastjson/parser/ParserConfig.java

This function is responsible for checking if a class specified by `@type` in a JSON string is allowed to be deserialized. Before the patch, it had insufficient validation and would load arbitrary classes, leading to Remote Code Execution. The patch adds several security checks, including verifying the class has a `@JSONType` annotation before loading it, and restricting which `expectClass` types are trusted.

com.alibaba.fastjson.util.TypeUtils.loadClass

src/main/java/com/alibaba/fastjson/util/TypeUtils.java

This utility function is used by `checkAutoType` to load a class by its name. The vulnerable version would cache loaded classes by default, which could be abused by an attacker. The patch changes the default behavior to not cache classes unless explicitly requested, and adds checks for class name length.

com.alibaba.fastjson.serializer.MiscCodec.deserialze

src/main/java/com/alibaba/fastjson/serializer/MiscCodec.java

This function handles the deserialization of special types, including `java.lang.Class`. The vulnerable version called `TypeUtils.loadClass` in a way that cached the loaded class. An attacker could use this to get a malicious class into the cache, potentially bypassing later security checks.

Vulnerability Intelligence
Miggo AI

Unlock WAF rules for this CVE

Generate vendor-ready rules for the observed attack patterns, plus reasoning and safe deployment guidance

Get WAF rules

WAF Protection Rules

WAF Rule

W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.

Reasoning

*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.