Exploit catalog results

Showing 3 PoCs on this page

GitHub

tr3m0x/CVE-2021-41773

Repository PoCStars: 0Created 2026-07-27
WriteupCVE-2021-417733 files

5.1 KiB

Analysisdeepseek-v4-pro:cloud ·

Technical assessment

The artifact is a technical writeup explaining the CVE-2021-41773 path traversal vulnerability in Apache HTTP Server 2.4.49. It includes a root cause analysis, a vulnerable code snippet, a processing flow diagram, and instructions for setting up a lab environment and sending a curl request to demonstrate the vulnerability. No executable exploit or scanner code is provided.

Backdoor review

No backdoor observed in reviewed code

The repository contains only a Dockerfile and a Readme.md documenting the CVE-2021-41773 path traversal vulnerability. The Dockerfile builds a vulnerable Apache 2.4.49 image and copies a configuration file (not included in the reviewed text). The Readme.md provides a technical explanation, root cause analysis, and a curl command to demonstrate the vulnerability. No backdoor, deceptive payload, or concealed executable behavior is present in the reviewed evidence.

ClassificationWriteup
Model confidence95%
AuthenticationNot required
Languagesmarkdowndockerfile
Target softwareApache HTTP Server
Attack typespath traversal
Evidence & reasoningClassification basis · observed behavior · safety review
Technical evidence

Classification basis and observed behavior

Classification basis

The artifact is a Markdown document containing a detailed technical analysis of CVE-2021-41773, including root cause, vulnerable code, and exploitation steps. It does not contain any executable code that automatically exploits or scans for the vulnerability; the provided curl command is a manual example for demonstration purposes. The classifier identified no PoC code.

Readme.md:1-307Dockerfile:1-5

Requirements

  • Apache HTTP Server 2.4.49 with Alias-like directives configuredReadme.md:5-7
  • Docker to build and run the vulnerable environmentReadme.md:223-234

Observed behavior

  • Describes how an attacker can use encoded dot sequences (e.g., .%2e) to bypass path traversal checks and read arbitrary files like /etc/passwdReadme.md:15-43
  • Provides a curl command to send a malicious request to a locally running vulnerable serverReadme.md:243-245
Safety-review evidence

Behaviors behind the backdoor verdict

Observables

Vulnerability Documentation
Payload withheldThe Readme.md explains the vulnerability, provides a PoC curl command, and discusses canonicalization issues. This is standard educational/exploit documentation.Readme.md:1-307
Docker Build Instruction
Payload withheldThe Dockerfile builds a container with the vulnerable Apache version for testing the PoC.Dockerfile:1-5
Review boundaries

What the analysis did not establish

  • One file (vulnerable.conf) is omitted from the evidence, so the exact Alias configuration is not visible.
  • The analysis is based solely on the provided text; no dynamic testing was performed.
  • The file 'vulnerable.conf' referenced in the Dockerfile is not included in the reviewed text evidence; its content is unknown.
  • Binary files were not analyzed per the evidence envelope policy; no binaries were 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.

GitHub

tr3m0x/CVE-2019-25065-poc

Repository PoCStars: 0Created 2026-03-26
ExploitCVE-2019-250652 files

2.9 KiB

Analysisdeepseek-v4-pro:cloud ·

Technical assessment

A Python script that exploits CVE-2019-25065, an OS command injection vulnerability in OpenNetAdmin <=18.1.1, by sending a crafted POST request to execute arbitrary commands on the target.

Backdoor review

No backdoor observed in reviewed code

The PoC consists of a Python script that exploits CVE-2019-25065 (OS command injection in OpenNetAdmin) and a short README. The script sends crafted POST requests to a user-supplied URL to execute an arbitrary command provided by the user. All behavior is consistent with a standard public exploit: it accepts a target URL and command, sends the injection payload, and prints the output. No concealed executable behavior, persistence, credential theft, unrelated remote access, or deceptive payloads were observed.

ClassificationExploit
Model confidence100%
AuthenticationNot required
LanguagesPython
Target softwareOpenNetAdmin
Attack typesOS command injection
Evidence & reasoningClassification basis · observed behavior · safety review
Technical evidence

Classification basis and observed behavior

Classification basis

The script constructs and sends a malicious HTTP request that injects an OS command into the 'ip' parameter, which is a classic exploit technique. It includes a function named 'exploit' that executes the payload and returns the output, demonstrating active exploitation rather than mere detection.

exploit.py:28-50Readme.md:3

Requirements

  • Target running OpenNetAdmin <=18.1.1 with the vulnerable endpoint accessible.Readme.md:3
  • Network access to the target URL.exploit.py:18

Observed behavior

  • Sends a POST request with a command injection payload in the 'xajaxargs[]' parameter to execute an arbitrary OS command.exploit.py:30-37
  • Extracts and prints the command output from the HTTP response using a unique separator.exploit.py:42-46
  • Optionally retrieves the OpenNetAdmin version by querying the app_about module.exploit.py:10-23
Safety-review evidence

Behaviors behind the backdoor verdict

Observables

Command Injection Payload
Payload withheldThe exploit constructs an OS command injection payload using a user-supplied command, delimited by a UUID separator to extract output. This is the expected exploit mechanism for CVE-2019-25065.exploit.py:34
Network Request
Payload withheldThe script sends HTTP POST requests to the target URL with the injection payload. This is the standard delivery method for the exploit.exploit.py:40
Review boundaries

What the analysis did not establish

  • The evidence does not include any authentication mechanism; the exploit appears to work without credentials, but the CVE context lists 'PR:L' (privileges required: low), which may indicate a discrepancy or that the exploit bypasses authentication.
  • Only the two text files (Readme.md, exploit.py) were reviewed; no binary files were present.
  • The behavior of the Python standard library and the 'requests' third-party library is assumed to be benign; their source code was not inspected.
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.

GitHub

tr3m0x/CVE-2025-6019

Repository PoCStars: 0Created 2026-02-13
ExploitCVE-2025-60193 files

4.5 KiB

Analysisdeepseek-v4-pro:cloud ·

Technical assessment

A bash script that exploits CVE-2025-6019 by creating a loop device for a crafted XFS image, triggering a UDisks2 filesystem resize via D-Bus, and then executing a SUID-root shell from the mounted image to achieve local privilege escalation.

Backdoor review

No backdoor observed in reviewed code

The supplied evidence consists of a README and a bash exploit script for CVE-2025-6019. The script performs a local privilege escalation by creating a loop device, triggering a UDisks2 filesystem resize, and executing a pre-staged SUID rootbash. No backdoor, deceptive payload, or concealed operator-directed harm was observed. The behavior is consistent with the stated exploit goal.

ClassificationExploit
Model confidence95%
AuthenticationRequired
LanguagesShell
Target softwarelibblockdevudisks2
Attack typesLocal Privilege Escalation
Evidence & reasoningClassification basis · observed behavior · safety review
Technical evidence

Classification basis and observed behavior

Classification basis

The script actively performs the steps to exploit the vulnerability: it sets up a loop device, triggers the vulnerable resize operation via D-Bus, and then executes a SUID-root shell to escalate privileges. This is exploitation, not just detection or scanning.

exploit.sh:32-64Readme.md:3

Requirements

  • An allow_active Polkit session (e.g., physical console login)Readme.md:13
  • A crafted XFS filesystem image containing a SUID-root shell, generated by payload withheldReadme.md:14
  • Required tools: mkfs.xfs, udisksctl, gdbus, loginctlReadme.md:12

Observed behavior

  • Verifies the user has an active session using loginctlexploit.sh:9-25
  • Kills the gvfs-udisks2-volume-monitor processexploit.sh:29
  • Creates a loop device for the provided XFS image using udisksctlexploit.sh:32-36
  • Starts a background process that repeatedly attempts to execute a SUID-root shell from /tmp/blockdev* to keep the filesystem busyexploit.sh:40-43
  • Triggers the UDisks2 filesystem resize operation via a D-Bus call to org.freedesktop.UDisks2.Filesystem.Resizeexploit.sh:46-49
  • Locates and executes the SUID-root shell (rootbash) from the /tmp/blockdev* directory to gain root privilegesexploit.sh:55-64
Safety-review evidence

Behaviors behind the backdoor verdict

Observables

Exploit Behavior
Payload withheldThis is the normal exploit behavior for CVE-2025-6019 as described in the README.exploit.sh:32exploit.sh:47-49exploit.sh:62-64
File Reference
Payload withheldThe README references a script to build the XFS payload image, but this file was not included in the supplied text evidence.Readme.md:7
Review boundaries

What the analysis did not establish

  • The payload withheld script, which builds the malicious XFS image, is not included in the selected text evidence.
  • The evidence does not include the SUID-root binary (rootbash) or the crafted XFS image itself.
  • The analysis is based solely on static review of the provided source code; the exploit was not executed.
  • The payload creation script payload withheld was not included in the supplied text evidence.
  • One file (unclassified, 1435 bytes) was omitted from the text evidence and not analyzed.
  • Binary content was not analyzed per the evidence envelope policy.
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.