Web Security · 15 Feb 2026
AST SQL Bypass in Modern WAFs
Analyzing the failure of Abstract Syntax Tree (AST) tokenization filters when confronted with context-aware, deterministic syntax mutations.
Read note ↗Research archive
Evidence, explained.
Long-form publications, focused research notes, and vulnerability disclosures—written to make the work reproducible.
Publications
0201↗02↗
WSHawk: Stateful Security Assessment of WebSocket Applications through Adaptive Payload Mutation and Browser-Assisted Validation
Most automated web-application scanners are built around the request/response model of HTTP and do not translate cleanly to WebSocket endpoints, where a single long-lived, bidirectional connection carries appli…
Basilisk: An Evolutionary AI Red-Teaming Framework for Systematic Security Evaluation of Large Language Models
Basilisk is an open-source artificial intelligence (AI) red teaming and large language model (LLM) penetration testing framework. It maps the adversarial attack surface of LLM applications to the OWASP LLM Top …
Research notes
03Web Security · 20 Jan 2026
WebSocket Threat Vector Analysis
Uncovering race condition patterns and blind server hijackings in active full-duplex WebSocket channels.
Read note ↗AI / ML · 10 Dec 2025
Genetic Prompt Mutations on LLMs
Applying genetic algorithms to automate safety evaluation vectors against boundaries in Large Language Models.
Read note ↗Disclosures
08code-graph-rag Approval Bypass via GNU find -exec
code-graph-rag <= 0.0.669
Read disclosure ↗lollms-webui Server-Side Request Forgery via /api/proxy
ParisNeo/lollms-webui
Read disclosure ↗lollms-webui Unauthenticated Remote Code Execution via Insecure eval()
ParisNeo/lollms-webui 20.0 Alpha
Read disclosure ↗Ollama GGUF String Length Panic Denial of Service
Ollama < v0.3.14
Read disclosure ↗Ollama Resource Exhaustion via Unbounded vocab_size
Ollama < v0.3.14
Read disclosure ↗LiteLLM Sandbox Escape via Unicode Normalization Bypass
LiteLLM < v1.51.5
Read disclosure ↗LiteLLM Server-Side Request Forgery inside Custom Guardrails
LiteLLM < v1.51.0
Read disclosure ↗LiteLLM Authentication Bypass via Pass-the-Hash
LiteLLM < v1.50.0
Read disclosure ↗