Tag: python
41 articles
ChocoPoC Malware Targets Vulnerability Researchers via Fake PoC Repos
Beware of fake proof-of-concept repositories on GitHub - a new malware called ChocoPoC is hiding in plain sight, stealing data from vulnerability researchers through a cleverly designed trap. This sneaky malware uses a dependency chain to infect systems, masquerading as a harmless Python proof-of-concept exploit.
Malware Exploits GitHub PoCs to Target Cybersecurity Researchers
Cybersecurity researchers are being targeted by a sneaky new campaign that uses malicious GitHub proof-of-concept exploits to deliver a remote access trojan, with over 2,400 downloads of a trojanized Python package already recorded. The attack unfolds through a multi-stage supply-chain trick involving compromised PyPI packages.
Malware Delivered via Trojanized GitHub Exploits Targets Security Researchers
Security researchers have been targeted by a sneaky malware campaign that uses trojanized GitHub exploits to deliver a Python-based remote access trojan, hiding in plain sight within popular proof-of-concept code repositories. The malware, downloaded over 2,400 times mostly on Linux-based systems, was spread through malicious packages cleverly concealed in dependency lists on GitHub.
Malicious PyPI Packages Expose Telegram Bot Servers to Hacker Control
Hackers have launched a sneaky attack, hiding malicious code in fake Python packages on PyPI, which can take control of Telegram bot servers and give attackers access to sensitive info like chats, contacts, and environment variables. This backdoor can be activated with a simple command, allowing attackers to execute any Python code on the victim's machine.
AI Code Review Foils Malicious Backdoor in Python Project
When Roman Imankulov analyzed a suspicious Python project with his AI agent, it quickly flagged a malicious backdoor, saving him from a potentially disastrous mistake. The AI code review proved to be a crucial safeguard, alerting Imankulov to walk away from the tainted code.
PyPI Packages Poisoned in Hades Supply Chain Attack
Malicious actors have launched a supply-chain attack on the Python Package Index (PyPI), infecting 19 packages with 37 tainted versions that can download and execute a hidden JavaScript payload. This sneaky Hades campaign uses poisoned Python packages to spread its reach, putting developers and users at risk.
Shai-Hulud Malware Targets Python Packages, Exposes Developer Secrets
Hundreds of thousands of downloads of 19 popular Python packages were compromised in a massive supply-chain attack that stole developer secrets, courtesy of the Shai-Hulud malware. The malicious packages, disguised as useful bioinformatics and science tools, were actually designed to expose sensitive information.
ChromaDB Flaw Enables Server Hijacking via AI Model Exploit
A newly discovered vulnerability, CVE-2026-45829, in ChromaDB's Python FastAPI variant allows hackers to hijack servers by exploiting AI models, with a security expert noting that authentication is present but poorly placed. This flaw lets unauthenticated attackers run arbitrary code on exposed servers by cleverly manipulating API endpoints.
LiteLLM SQL Flaw Exploited 36 Hours After Disclosure
A critical SQL injection flaw, CVE-2026-42208, was exploited just 36 hours after its disclosure, putting vulnerable LiteLLM versions at risk of unauthorized database access. The bug, with a CVSS score of 9.3, allows unauthenticated callers to reach a vulnerable database query through the proxy's error-handling path.
Unpatched Flaw Exposes Hugging Face LeRobot to Remote Code Execution
A critical, unpatched vulnerability in Hugging Face's LeRobot platform, rated CVSS 9.3, allows hackers to remotely execute code by exploiting Python's insecure pickle format, putting users at risk of devastating attacks. This flaw enables unauthenticated attackers to gain control by deserializing malicious data sent over unsecured channels.
Hackers exploit Marimo flaw to spread NKAbuse malware via Hugging Face
Hackers are exploiting a critical flaw in Marimo's reactive Python notebook to spread a new variant of NKAbuse malware, sneaking malicious payloads onto Hugging Face Spaces, a popular platform for sharing machine learning models. This alarming attack highlights the need for vigilance when it comes to defending against malware disguised as code-sharing tools.
Marimo Flaw CVE-2026-39987 Exploited Rapidly After Disclosure
A single line of code can drastically change the risk landscape for thousands of users - and that's exactly what happened with Marimo, an open-source Python notebook, when a critical vulnerability (CVE-2026-39987) was exploited just 10 hours after its disclosure. This severe flaw, with a CVSS score of 9.3, allows pre-authenticated remote code execution, putting all Marimo versions prior to the disclosed fix at risk.
ComfyUI Instances Enlisted in Widespread Cryptomining Botnet Campaign
A sneaky campaign is on the hunt for exposed ComfyUI instances, using them to fuel a cryptomining botnet and secretly install malicious nodes - putting unsuspecting users' systems at risk. This covert operation uses a Python scanner to scour cloud IP ranges, exploiting vulnerabilities and turning systems into cryptocurrency-mining machines.
PickleScan Exclusive: Critical Flaws Rock AI Supply Chains
Researchers disclosed three critical PickleScan zero-days that let attackers stealthily swap or tamper with local AI models—injecting misinformation, bias, or even exfiltrating data from Python/PyTorch model runners. Exploitable via drive-by browser-origin attacks against assumed-safe local admin endpoints, these flaws show how our trusted AI tooling can become the weakest link in the supply chain.
Legacy Python Bootstrap Scripts: Stunning PyPI Threat
Legacy zcbuildout scripts left in projects can become silent attack vectors—if a referenced domain lapses and an attacker reclaims it, builds can pull and execute malicious code that reaches PyPI. ReversingLabs’ findings show how a tiny oversight in old bootstrap helpers can enable wide supply‑chain compromise, so it’s time to find, update, or remove those scripts.
Vulnerable Rust crate: Stunning critical uv Python flaw
async-tar, a tiny Rust crate, unexpectedly sparked a chain reaction when a flaw in a forked copy rippled into fast uv, showing how fragile ecosystems built on forks can be; one fork is patched, but the most widely downloaded release still sits unpatched.
Vulnerable Rust crate Exclusive: Critical uv Python Flaw
If you use uv Python, take note: a critical flaw in the Rust crate async‑tar was patched in one fork, but the most widely distributed uv build still ships the vulnerable copy. It’s a clear reminder that fixing one fork doesn’t secure an ecosystem built on cloning and convenience.
Python backdoors: Exclusive Risky Threat Warning
Researchers warn the Confucius espionage group is shifting from weaponized documents to Python backdoors like AnonDoor, widening the attack surface and making detection much harder. Organizations should boost visibility into scripting, enforce least privilege, and monitor package and repository activity before attackers hide in legitimate developer tooling.
phishing campaign: Risky PyPI Scam — Must-Read Alert
Got an email asking you to verify your PyPI credentials? Change your password and enable MFA right away — attackers are running a convincing fake PyPI site to harvest logins and could use stolen accounts to push malicious packages or compromise your supply chain.
PyPI packages: Risky SilentSync Alert — Must-Have Fix
Cybersecurity researchers found two malicious PyPI packages that delivered the SilentSync RAT to Windows machines, enabling remote command execution, file theft and screen capture. Treat your dependency tree like an attack surface—audit packages, pin versions and lock down CI to stop supply-chain intrusions.
AI-native Villager: Risky Exclusive Tool Sparks Alarm
A China-origin tool called AI-native Villager has quietly topped 11,000 PyPI downloads, combining Kali Linux and DeepSeek into an easy-to-use pen-testing automation that’s as useful for defenders as it is tempting for attackers. That rapid uptake underscores a growing dilemma: powerful, AI-driven tooling can speed security work — and just as quickly widen the pool of potential abusers.
CastleRAT malware: Exclusive Dangerous C/Python Threat
A new strain of CastleRAT, now rewritten in both C and Python, is being spread via a nasty ClickFix trick that convinces users to paste malicious commands into their terminals—don’t paste commands you don’t trust. Stay skeptical of unsolicited “fixes,” verify sources, and treat pasteable commands like executable attachments.
Alarming 188% Annual Increase in Malicious Open Source Packages
Discover the shocking 188% annual rise in malicious open source packages and its implications for developers and software security.
Over 200 Compromised GitHub Repositories Discovered in Attack on Gamers and Developers
Over 200 compromised GitHub repositories linked to a cyber attack targeting gamers and developers, raising security concerns in the tech community.