1st Place
blartclaw
building AI surveillance agents for security in a city setting.
securityvisionagents
Aiden Cohen, Kris Diallo, Brendan Fay
AI lab hackathons, 2025 to now
What got built, and what won.
Every submission from hackathons run by Anthropic, OpenAI, Google, Meta and the other big labs, with the winners marked and the code linked.
- projects
- 31,980
- hackathons
- 93
- award winners
- 585
- inferred from GitHub
- 433
GitLab - 1st Place
mnemosyne
The AI that remembers every failure — predicts outages, annotates risky code inline, gates CI pipelines, auto-blocks critical MRs, and diagnoses failures. Powered by Gemini and gitlab on Google Cloud.
dev toolssecurity
2nd Place | Developer Tools
Sentinel
Catch security holes in your MCP server before you ship it. Static rules, GPT-5.6 review, and Docker-sandboxed probes-every finding mapped to the OWASP Agentic Top 10.
securitydev tools
Malik Bashaar Javaid
GitLab - 2nd Place
STYLOMETRY
STYLOMETRY: The forensic guardrail for your GitLab pipeline
dev toolssecurity
Anna Tchijova, Olga Vasilieva
2nd Place
ShareLock
ShareLock is an AI-assisted reconstruction tool that helps turn fragmented witness memories into visual leads.
securityvision
Max Neuwinger, Dalia Shaaban, Michal Tesnar
2nd Place
S3 Butler
S3 Butler provides analysis of S3 buckets, visualizations of storage traffic, and visibility into user permissions.
dev toolssecurity
Anurag Mittal, Taylor McKinnon
Parallel - 2nd Place
Consentinel
Consentinel: one registry, two directions hunting unauthorized AI clones of performers worldwide in five languages, clearing studio deliveries before ship. AI finds evidence; humans decide.
legalagentssecurity
Prachit Tupe, Vedant Devkar, Swara Vedak
Arize - 2nd Place
ComplianceOS
A compliance agent that audits your code against real law, writes the fixes, and reads its own Phoenix traces through MCP to get better after every scan.
legalsecuritydev tools
Bhupen Sai Bandi, Aryaman Mishra
2nd Place
1. Computer Use Attack
1.
securityagents
Yonggyu Kim
Elastic - 3rd Place
Carryover
Visa intelligence that protects migrants from agency fraud. Three AI agents on Elasticsearch + Gemini that don't just inform but they detect scams in real time and file real complaints.
securityagents
Yordanos Kassa
Dynatrace - 3rd Place
Heimdall: Autonomous Incident Response for Dynatrace
Autonomous SRE that watches your Dynatrace tenant 24/7, diagnoses production incidents from real spans and Davis CoPilot, and opens a pull request before your phone stops ringing.
dev toolsagentssecurity
3rd Place
Codeminder
Codeminder is a hybrid LLM-based SAST (Static Application Security Testing) tool.
securitydev tools
Agajan Torayev
20 Category Winners
SafeRoute Pune – Women's Safety Route Mapping System
An AI-powered safety-first navigation web app that helps women in Pune choose the safest walking routes using Google Maps data, safety zones, and real-time risk analysis.
security
Vaishnavi Ghuge, Arpita Kulkarni, Komal Puranik +1
Best Use of NemoClaw
MonkeyClaw
MonkeyClaw: Probing, patching, and protecting the NVIDIA NemoClaw ecosystem.
securityagents
Cofee Coder, ezzy1630 Rappeport, Justin Lee
Honorable Mentions
Proofy.AI
An explainable, Gemini-powered verification platform built to restore trust in digital media.
securityvision
Jivitesh reddy, Nathan James, Adhidev Jayaraj
Honorable Mentions
Vigil AI
Vigil AI: Securing financial microservices with autonomous, real-time fraud mitigation.
securityfinanceagents
Ayan Liger
Honorable Mentions
PROCSee
PROCSee turns your system into a crime scene — and autonomously investigates every process
security
Mir Abbas
WebMCP Challenge Winners
Mandate
Mandate turns updatable fields into temporary mandates — a live WebMCP tool compiled from what one person chose, expiring on its own, and re-checked by the server on every call.
securityagents
Harzer Heribert
Best Architectural Design
Helixstreet — autonomous genomic fleet
Drop raw DNA in a bucket and walk away: five Gemini+Gemma agents screen PII, call variants with DeepVariant, flag drug-gene risks, and file an auditable digital twin. No chat. No clicks. No human.
healthagentssecurity
Thomas Deisen
Winners
SafeScreen AI
An on-device visual safety layer that helps protect people from explicit, abusive, and manipulated media before they interact with it.
securityvision
safescreen ai
Runner-up
Enamel
The AI operating system for a small dental clinic — records, scheduling, billing, lab and stock. The AI runs on the clinic's own computer, fully offline. One-time licence, in daily clinical use.
healthsecurity
Yatin Shinde
API Credits + Cash
TerpWalk
An AI-powered website that helps women find the safest route to get from point A to point B at any time of the day.
securityagents
Somya Kochar, Disha Tulsyan
Honorable Mention: Most Creative Use of Claude
Guardian
Guardian watches every page your grandparent visits, speaks up about scams, and walks them through tasks one step at a time - no typing required.
accessibilitysecurityagents
Boris Hong Pak Lo, Jay Thapar
Honorable Mentions
AgentGuard: The Semantic Firewall for the Agentic Web
The SSL for AI Agents. A Semantic Firewall that uses Gemini 3.0 to distinguish helpful shoppers from malicious scrapers in real-time.
securityagents
Mayur Pawar
Best of AI Agents
PermitScan AI
PermitScan Ai automates hazard discovery → permit creation → permit validation → permit refinement for oil & gas Work Orders (WOs).
agentssecurity
Sagar Bansal
Top 5
AI Security Engineer
Plug and play AI security engineer - helps you identify security vulnerabilities in the code and also generate patches to fix them. Always on security - now possible!
securitydev tools
Anupam Mediratta
Apple Mac Mini M4 (per team member)
Voice Guardian
Protecting Your Home with AI Voice
securityvoice
Yuji Kwon, Jamie Kim, Changho Choi +1
Finalist
SecureLoop
SecureLoop turns production Sentry alerts into Codex-generated, sandbox-tested, human-approved fixes inside JetBrains.
securitydev tools
Abhiram Yadu Sribhashyam, Peyton Li, Rahul Marri
Finalist
YBIGTA2026
프로젝트 Elder Guard는 기존의 불안정한 에이전트 시스템 위에 제어 레이어와 안전 검토 계층을 추가해, 노인도 쉽게 사용할 수 있는 < 사기 방지 디지털 안전 에이전트 > 를 만드는 것을 목표로 합니다.
securityaccessibilityagents
Heechan Yun, Sungwoo Kang
TaterGang
Plannotator Live (built from scratch at the hackathon) is a collaborative planning review surface for agentic engineering; built for the world where coding agents are fast but planning and review have become the new bott
dev toolssecurity
Michael Ramos
Agent Beacon
Agent Beacon is a CLI for real-time agent safety and observability.
securitydev tools
Souvik Banerjee, Justin D'Souza
AgentShield
AgentShield scans AI agent configs for security vulnerabilities.
securitydev tools
Affaan Mustafa
SkillStudio
A verified registry + CLI for Gemini agent skills.
securitydev tools
Anton Abyzov
VinceOvando
LLMitM v2 is an autonomous pentesting agent where Opus 4.6 builds the automation that replaces itself.
securityagents
Vince Ovando
Proxie
Spectra deploys 8 Claude Opus 4.6 agents — 6 parallel specialists, a MetaPrompter, and a CritiqueAgent with Extended Thinking — to analyze entire codebases in under 5 minutes.
dev toolssecurity
Vivek Kumar
Tortilya
In five days, one researcher and Claude Opus 4.7 broke the most capable LLMs on the planet — twenty-four different ways — then turned every failure into a lesson, a defense, and a public certification course.
securityeducation
Ilja Nevolin
Ryan MacDonald
Millions of servers run the world's critical infrastructure, defended by teams too small to keep up.
securityagents
Ryan MacDonald
AI GitHub Auditor - Project Automation Assistant
An AI-powered assistant that audits GitHub repositories, detects code quality and security issues, and generates human-readable reports and tracking project metrics.
dev toolssecurity
OpenRange
OpenRange is a multi-agent cybersecurity gymnasium where Red and Blue agents co-evolve on validated enterprise networks that mutate between episodes, requiring long-horizon planning across 50–100 step trajectories with s
securityagents
Anto Joseph, Lars Talian Stangebye-Hansen, Aaron Brown
Airlock
Airlock is an open-source, cryptographic context broker for AI agents.
security
Shukan Shah
Heimdall
Anthropic does not manage or audit MCP servers.
securitydev tools
Marwane Chahed
Bears
Our project is an environment designed for an agent to oversee 7 agents playing a popular strategy game Diplomacy.
agentssecurity
Masaki Tanaka Allwardt, Ronok Tanvir, Aryan Gupta
Techulus
A fast, offline security scanner that protects AI agents from prompt injection, credential leaks, and code injection attacks, the security layer every AI agent needs.
securityagents
Arjun Komath
ForwardCheck
Your mom forwards you a message.
agentssecurity
Chandra mohan Mohan
AgentPass
AgentPass is the identity layer for autonomous AI agents — "Auth0 for AI Agents." Today, agents (Claude Code, CrewAI, AutoGPT) can't independently register on websites, verify via email/SMS, or manage credentials.
securityagents
Roman Kudin
Geoffrey Young
Claude Code plugins inject content into Claude's reasoning via the hook system's exit-2 stderr channel, unfiltered and unattributed.
securitydev tools
Geoffrey Young
Wi-Fi Kids - Intelligent Parental Control
Wi-Fi Kids turns the router into a digital tutor: before going online, kids answer adaptive quizzes powered by GPT-5. A parental control that educates while limiting excessive screen time.
educationsecurity
tophawkteam
Pythia
Pythia is a Claude-powered oracle agent that intervenes in real time during malware emulation.
security
Lukas Machado
Regula
Regula is a self-driving NIS2 compliance audit on Claude Opus 4.7.
securitylegal
Maciek Lagwa
Mesmer
Mesmer is an open-source cognitive red-teaming toolkit for LLM applications.
securityagents
Galih Laras Prakoso
Vectimus Warden
Vectimus Warden is the containment layer for AI agents, from Claude Code sessions to autonomous fleets on Amazon Bedrock AgentCore.
securityagents
Joe Holland
WatchDog
WatchDog is an RL environment training AI oversight agents to detect errors in real-time.
securityagents
Mooizz Abdul, Abhilash Ganga, Raghu Vamshi Hemadri
SiteGuard AI
SiteGuard AI는 단순히 현장의 인원이 몇 명인지 세는 것을 넘어, **"특정 인원이 이 공간을 이용할 때의 위험을 사전에 예측"**하는 차세대 군집 안전 시뮬레이션 시스템입니다.
securityvision
SK Kim
Forge Proof
Forge Proof combines Claude Opus with formal verification to find smart contract vulnerabilities with mathematical proof.
securityagents
Anto Joseph
Ethos Academy
Ethos Academy: Where your AI agents learn integrity, logic, and empathy.
securityagents
Allie Jones
Sanches
Sanches is a desktop app designed to enhance the software developer experience by analysing their project code in real-time to identify security vulnerabilities, notifying the developer and classifying vulnerabilities ac
securitydev tools
Dennison Lau, Anuroop Keshav, Nasimi Mamedov +1
BUZZ
⚡ Buzz is an autonomous, codebase security auditor/mapper and Pull Request remediator powered by Gemini 4 Argon’s 1M-token context window.
securitydev tools
James Orellana, Nisarga Patel
ARGUS
ARGUS: Autonomous Security Orchestrator ARGUS is an autonomous multi-agent security orchestrator.
securitydev tools
Chaitanya Uge
Rakshak
Rakshak is a proactive AI intervention system that transforms AI from a passive assistant into an active digital guardian.
securityfinancevoice
Abhinandan Mishra, Satya Sri Dheeraj Motupalli
Rakshak
Rakshak is a real-time AI conversation guardian that protects users from cyber fraud such as digital arrest scams, UPI fraud, OTP phishing, and social engineering attacks.
securityvoice
Sai Nitish Mitta, Mittapalli Hareesh, Shubham Gupta +1