Marcello Maugeri
  • Bio
  • Experience
  • Publications
  • Projects
  • Team
  • Blog
  • Projects
    • AMAB: Adversarial Manoeuvre Attack Benchmark
    • BigEye
    • Automated Testing of GraphQL APIs (ATG)
    • α
    • Eulibra
    • PolyMirror.AI
    • LibAFLstar
    • TonAIStark
    • Flare-FL
    • Proteus
    • Fuzzing Matter(s)
    • PECS - Privacy Enrooted Car Systems
    • The Water Code
    • Fantamaniaci (iOS)
    • Dragomanni Apartments Website
    • Do a Shot (iOS)
    • Distantia
  • Publications
    • {WIP}: A Benchmark for Security Evaluation of Adversarial Manoeuvre Attacks in Autonomous Driving
    • RTE4SDC at the ICST 2026 Tool Competition—Self-Driving Car Testing Track
    • Automation Challenges and Solutions in Coverage-Guided Fuzzing of Multiprocess Software Systems
    • KrakQL: LLM-Guided Blind Introspection of GraphQL Schemas
    • BenGQL: An Extensible Benchmarking Framework for Automated GraphQL Testing
    • Fuzzing OPC UA with AFLNet, ChatAFL and LibAFLstar: A Research Experience Paper
    • AI-related Vulnerabilities within CVEs: Are We Ready Yet? A Study of Vulnerability Disclosure in AI Products
    • LibAFLstar: Fast and State-Aware Protocol Fuzzing
    • Poster: Machine Learning for Vulnerability Detection as Target Oracle in Automated Fuzz Driver Generation
    • Privacy Enrooted Car Systems (PECS): Preliminary Design
    • User-Empowered Federated Learning in the Automotive Domain
    • Wendigo: Deep Reinforcement Learning for Denial-of-Service Query Discovery in GraphQL
    • Fuzzing Matter(s): A White Paper for Fuzzing the Matter Protocol
    • Forkfuzz: Leveraging the Fork-Awareness in Coverage-Guided Fuzzing
    • Evaluating the Fork-Awareness of Coverage-Guided Fuzzers
    • Embedded fuzzing: a review of challenges, tools, and solutions
  • Upcoming Talks
    • BenGQL: An Extensible Benchmarking Framework for Automated GraphQL Testing
    • AI-related Vulnerabilities within CVEs: Are We Ready Yet? A Study of Vulnerability Disclosure in AI Products
  • Projects
  • Experience
  • Hound Team
  • Teaching
    • Learn JavaScript
    • Learn Python

Proteus

25 Oct 2024 · 1 min read

Proteus is an LLM-powered lending agent for the Aave V3 protocol designed to automate and optimise lending strategies. The project won 1st Place in the RNDM Agent Track at the Encode London 2024 Hackathon.

Last updated on 25 Oct 2024
AI Blockchain
Marcello Maugeri
Authors
Marcello Maugeri
Security Research Engineer, PhD
← Flare-FL 9 Feb 2025
Fuzzing Matter(s) 23 Feb 2024 →

© 2026 Marcello Maugeri. This work is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.