Stars
Terminal operations deck for independent model swarms. Freeze the target, hunt in private workspaces, seal each answer, and blind the judge so agreement cannot turn a copied write-up into a discovery.
AI/LLM-assisted JS Recon and vulnerability triage for authorized bug bounty testing & pentesting — scope-gated, passive-by-default, human-verified
Reports from automated security audits made by Swival.
AI threat-model review agent — turns architecture docs, DFDs, and code into evidence-checked findings, a go-live decision, and an interactive attack-chain diagram. Runs on the Claude API.
Abliteration without the weights — projective refusal steering as a GLP vector + fail-closed boot hotfix.
An evolving repository of CloudTrail events with detailed descriptions, MITRE ATT&CK insights, real-world incidents, references and security implications
A method for detecting **insecure-by-design** flaws in cloud services through AI-assisted design review — architectural flaws that violate expected security guarantees. This directory holds the tax…
WhatAboutSAM is my custom PoC of a Windows SAM dumper
Draft a post once and format it for LinkedIn, X, Bluesky, Threads, Mastodon, Facebook, and Instagram — each platform's length and formatting limits applied live. Plus a LinkedIn-only browser extens…
Protocol-Introspecting Chunk Adapter for Resilient Data
Modified BloodHound.py CE ingestor for red team use. Obfuscates LDAP through python ldapx-py package, paces LDAP searches and host RPC, and resumes a stopped collection. Writes BloodHound CE JSON.
A CI/CD Red Team Framework for demonstrating Build Pipeline security risks.
Reverse engineer anything with agents, from app behavior down to native binaries.
Agentic AI security scanner that hunts exploitable vulnerabilities like an adversary, proves them with executable PoCs, and fixes them test-first. A maintained fork of Capital One's VulnHunter, reb…
AI SecOps Evaluation Framework: practitioner-led evaluation of AI-enabled technology and MDR services.
Content-blind reverse proxy for exposure-protected security scanning
Open taxonomy of 168 AI/LLM attack vectors with versioned, source-pinned relations to the OWASP LLM, Agentic and MCP Top 10s and MITRE ATLAS — ESF Phases 1-3 reference implementation
Uncensored AI models or those fine-tuned for cybersecurity tasks.
The HTTP/2 CLI for finding tiny timing differences and winning request races
Passive sniffing tool for capturing and visualising WiFi location data disclosed by iOS devices
BSSID Search in Apple's public WPS service
Research dump on how IOS Location Services work
Refusal removal as a loadable KV-cache graft — zero weight modification, no refusal-direction projection, fully reversible.
Claude Code sending messages locally to active sessions
Ask the Web Account Manager (WAM) for Entra ID tokens
An automated tool to test AI models against scope manipulation (deceiving an AI agent about its real target).
A modular, stack-agnostic toolkit for AI coding agents to autonomously find, reproduce, and patch vulnerabilities.