Ak-cybe/CVE-2025-68664-LangGrinch-PoC

GitHubRepository PoCScanner

PoC files

5 files

File viewing is interactive and short-lived. Downloads are password-protected ZIP archives using password eip.

GitHub
Analysisdeepseek-v4-pro:cloud ·

Technical assessment

The artifact is a payload generator and documentation for CVE-2025-68664. The Python script (langgrinch_fuzzer.py) generates JSON payloads for testing the vulnerability but does not contain code to send them to a target or exploit a live system. The README provides a technical writeup and instructions for using the generator. The primary operation is generating and listing test payloads, which is characteristic of a scanner/validator, not an exploit.

Backdoor review

No backdoor observed in reviewed code

The repository contains a documented proof-of-concept (PoC) for CVE-2025-68664, a LangChain serialization injection vulnerability. The supplied text files (README.md, langgrinch_fuzzer.py, requirements.txt) describe the vulnerability, provide payload examples, and implement a payload generator. No concealed executable behavior, operator-directed harm, persistence, credential theft, or unrelated payloads were observed. The code generates and displays JSON payloads for authorized testing; it does not execute them against any target. The artifact is consistent with a legitimate security research PoC.

ClassificationScanner
Model confidence95%
AuthenticationNot required
LanguagesPythonMarkdown
Target softwarelangchain-core
Attack typesDeserialization InjectionSecret ExtractionSSRFRemote Code Execution
Evidence & reasoningClassification basis · observed behavior · safety review
Technical evidence

Classification basis and observed behavior

Classification basis

The artifact's primary executable component (langgrinch_fuzzer.py) is a payload generator that creates JSON strings for testing. It does not contain any code to send these payloads to a target, trigger deserialization, or otherwise exploit a vulnerability. The README describes the vulnerability and provides a version detection script. This is consistent with a scanner (code that checks for a vulnerability without exploiting it).

langgrinch_fuzzer.py:1-406README.md:420-464

Requirements

  • A target application using a vulnerable version of langchain-core (< 0.3.81 or >= 1.0.0 < 1.2.5) that deserializes user-controlled data containing the 'lc' key.README.md:164-170

Observed behavior

  • Generates JSON payloads containing the 'lc' marker for secret extraction, SSRF, RCE, and file read attacks.langgrinch_fuzzer.py:54-240
  • Lists available payloads by category and exports them to a JSON file.langgrinch_fuzzer.py:346-381
  • Provides a version detection script to check if the installed langchain-core is vulnerable.README.md:420-464
Safety-review evidence

Behaviors behind the backdoor verdict

Observables

Vulnerability Poc
CVE-2025-68664 LangChain deserialization injectionThe entire repository is a PoC writeup and payload generator for the stated CVE.README.md:1-626langgrinch_fuzzer.py:1-406
Payload Generator
langgrinch_fuzzer.pyA Python script that generates and displays JSON payloads for the vulnerability. It does not execute the payloads or connect to any remote service.langgrinch_fuzzer.py:1-406
Author Attribution
Amresh Kumar (Ak-cybe)The author is identified in the README and source code. This is metadata and not a backdoor indicator.README.md:607langgrinch_fuzzer.py:9
Review boundaries

What the analysis did not establish

  • Two files (PAYLOADS.md and LICENSE) are omitted from the evidence, so the full payload arsenal and license terms are not available for review.
  • The evidence does not include any execution logs or network traffic to confirm the payloads work as described.
  • Two files (PAYLOADS.md, LICENSE) were present in the repository but not included in the text evidence; only their metadata was provided. No binary or non-text content was identified in the artifact.
Model interpretation

This review is limited to the supplied PoC code and context. It does not assert that the code works or is safe to execute.

Linked vulnerabilities

1