
chub-supply-chain-poc
Silent dependency injection through AI documentation pipelines. 240 isolated Docker runs proving Context Hub's zero-sanitization MCP server lets…

Silent dependency injection through AI documentation pipelines. 240 isolated Docker runs proving Context Hub's zero-sanitization MCP server lets…

Syntactic Ghost: An Imperceptible General-purpose Backdoor Attacks on Pre-trained Language Models

Embed multiple secret messages in LLM chat token choices using arithmetic/Discop steganographic coders, with bit-exact decoding and steganalysis…

Gemini - Production LLM Runtime Alignment Context Injection

Prohibited Items Segmentation via Occlusion-aware Bilayer Modeling (ICME 2025)

Research code for training adapters that make LLM agent backdoors survive benign fine-tuning, with trigger-rate scoring, SWE-bench evaluation, and…

Proof of concept for CVE-2024-24590

A novel adversarial attack on LLM based on the Exponentiated Gradient Descent technique.

Sentinel detection lab for MCP attack chains: CVE-2026-26118 SSRF token theft, tool poisoning, cross-server exfiltration, identity post-exploitation.…

Benchmark harness measuring where prompt injection defenses fire in tool-using LLM agent pipelines, tracking canary tokens across exposed, persisted,…

Research implementation of the Black-Hole Attack that poisons vector database indexes via centrality-driven hubness, with FAISS pipelines and…

Offline SAST/DAST triage engine that parses source code and replays HAR traffic to detect business-logic flaws, BOLA, race conditions, and privilege…

AI-powered reactive website defense system that detects attacks, analyzes them, and autonomously patches source code in real-time using LLM agents.

Embeds and detects keyed, style-based watermarks in LLM-generated Python, Java, and C++ code via CST rewriting, preserving functional correctness and…

Research code, benchmark and defense for target rebinding attacks that trick LLM safety guards into grading forged decoy content, plus segment-max…

Research implementation of Hop-Decayed Influence (HDI) and the 3S attack framework, exposing structural auxiliary indexing vulnerabilities in…

A testing framework to identify and demonstrate deserialization vulnerabilities in LangChain Core (<0.3.81). Educational use only

Find code that runs without you asking: repo files that execute on folder open (VS Code tasks, agent hooks, MCP, devcontainers) and Python…