Alternatives
Products that do what A CTF challenge that let's you attack an LLM-based agent does
Hello HN, given the recent interest in the security vulnerabilities of agentic LLM apps, we would love to hear everyone's thoughts on a Capture The Flag (CTF) challenge we built. The challenge is designed to test the robustness of LLM-based agent systems, showcasing how an attacker can extract sensitive information from the system. The challenge is part of the Invariant Summer '24 CTF series, where participants can play the challenge and compete for a prize pool of $1000. Let us know what you think and what attacks you tried. We are also open to any feedback on how we can improve with the…
- 1RU
hey all, happy to share research i've been working on for islo.dev in recent months. ever since the cheating agents (https://debugml.github.io/cheating-agents/) paper came out, revealing reward hacking was 4x more prevalent than previously estimated, i've been looking into how we can deal with the issue the common approach (taken by the tbench team) is post hoc trajectory analysis. i've been interested in the idea of reframing the problem as an endpoint security problem and tackling it via sandbox i hope you find it interesting, and thanks to the islo.dev team for…
Jun 2026 · github.com
- 2IS
Hey HN! For that last 8 months I've been trying to make agents that can hack web applications to find vulnerabilities in them - An AI Security Tester. The system has 29 agents in total, a custom LLM Orchestration framework which works on the task-subtask architecture (old-school but works amazingly for my use case, and is pretty reliable) with custom agent calling mechanism. No Auo-Gen, Langchain and Crew AI - Everything custom built for pentesting. Each test runs in an isolated Kali linux environment (on AWS Fargate), where the agents have full access to the environment to undertake any…
2025
- 3AR
If you're interested in exploring what LLM-based agent systems these days actually do to solve certain benchmarks such as SWEBench or WebArena, we created a small leaderboard with our team, that allows to view a lot of public and OSS agent results including all the runtime traces (the step-by-step reasoning behind the scenes). Looking at traces is actually quite interesting, as they reveal a lot about the inner working and shortcomings of current agent system, e.g. see https://explorer.invariantlabs.ai/u/invariant/webarena--SteP... for an example trace.
2024 · explorer.invariantlabs.ai
- 4OS
Hello HN, I’ve been building AI agents lately and ran into a common "Context Bloat" problem. When an agent has 20+ skills, stuffing every system prompt, reference doc, and tool definition into a single request quickly hits token limits and degrades model performance (the "lost in the middle" problem). To solve this, I built OpenSkills, an open-source SDK that implements a Progressive Disclosure Architecture for agent skills. The Core Concept: Instead of loading everything upfront, OpenSkills splits a skill into three layers: Layer 1 (Metadata): Light-weight tags and triggers (always loaded…
Jan 2026
- 5WM
Hi Hacker News! My friend and I [1] would like to share with you something we did this week-end. We really like the idea of game jams like the Ludum Dare, and we’d like to participate, but it’s often complicated for us, as we are often limited by the rules inherent to this kind of competition: we have to be available at specific dates, and the rules doesn’t always work for us. For example: if you want to participate as a team in a Ludum Dare compo, you have 72h instead of 48h, and you are not required to publish your code. We realized that what we really like is: - To work together. - The…
2014
- 6TA
Hello HN, I am Brian Cardinale, a penetration tester and security researcher at SecureCoders. We have been performing more and more AI based security assessments. We were presented a unique challenge of testing a system where the only interface was voice based, and as much as I like talking on the phone , we decided to create a test harness to facilitate the actual testing in a more systematic way. The technical test harness was the easy part, though. Creating test goals and attack strategies to help facilitate repeated and comprehensive testing became the real challenge. As such, we have…
Feb 2026 · redcaller.com
- 7AG
I've created a social deduction game for LLMs, in which the bots attempt to hunt each other. It's a Mafia group turing test: the models are told to find who the bot is - where, in fact and unbeknown to them, they are all bots. I did this a while back so models aren't the newest, and they are all non-thinking (for speed and token costs). Et voilà.
Jan 2026 · hiding-robot.vercel.app
- 8IB
Hi HN, I’m the creator of Cordum. I’ve been working in DevOps and infrastructure for years (currently in the fintech/security space), and as I started playing with AI agents, I noticed a scary pattern. Most "safety" mechanisms rely on system prompts ("Please don't do X") or flimsy Python logic inside the agent itself. If we treat agents as autonomous employees, giving them root access and hoping they listen to instructions felt insane to me. I wanted a way to enforce hard constraints that the LLM cannot override, no matter how "jailbroken" it gets. So I built Cordum. It’s an open-source…
Jan 2026 · github.com
- 9CB
AI agents now have impressive reasoning capabilities. This raises an important question: how dangerous are these AI agents at identifying & exploiting web vulnerabilities? We created CVE-bench to find out (I'm one contributor of 16). To our knowledge CVE-bench is the first benchmark using real-world web vulnerabilities to evaluate AI agents' cyberattack capabilities. We included 40 CVEs from NIST's database, focusing on critical-severity vulnerability (CVSS > 9.0). To properly evaluate agents’ attacks, we built isolated environments with containerization and identified 8 common attack…
2025 · github.com
- 10AP
Hi HN, I’m a solo developer and built AgentWatch to solve a problem I kept running into while building AI agents: preventing runaway loops and unexpected LLM spend before requests reach the model. AgentWatch sits in front of OpenAI, Anthropic, Gemini, Bedrock, Azure OpenAI, Groq, and others to enforce budgets and runtime policies. I’d really appreciate your feedback. If you’re building AI agents, does this solve a problem you’ve experienced? I’d also love to hear what you’d improve or challenge.
Jun 2026 · agent-watch.dev
- 11PP
I've been working on applying LLMs to long-context, verifiable problems over the past year, and today I'm releasing a benchmark of 62,000 pencil puzzles across 94 types (sudoku, nonori, slitherlink, etc.). The benchmark also allows for intermediate checks /rule breaks for all varieties at any step. I tested 51 models against a subset (300 puzzles) in two modes: single-shot (output the full solution) and agentic (iterate with verifier feedback). Some results: - Best model (GPT 5.2@xhigh) solves 56%. (~ half the puzzles are unsolved by any model) - Agentic solves average 29 turns. The…
Mar 2026 · ppbench.com
- 12HW
Hello everyone! I’m thrilled to announce the latest feature from Mutahunter.ai, the ultimate tool for finding and fixing weaknesses in your code. We’ve designed Mutahunter to leverage mutation testing powered by advanced LLMs, helping you uncover vulnerabilities and enhance your code quality effortlessly. Introducing our newest feature: Detailed Mutation Testing Reports! After running our mutation tests, Mutahunter now generates comprehensive reports that clearly summarize: • Vulnerable code gaps • Test case gaps These reports significantly reduce the cognitive load on developers by…
2024 · github.com
- 13IB
I built a tool to roast landing pages with AI agents. I was gathering feedback from watching landing page roast videos, and figured out I could prompt LLMs to analyse a screenshot and roast based on the same criteria. It's not 100% accurate yet, but it has been really insightful when I've tested it on my own websites. Let me know what you think!
2024 · roastmylandingpage.io
- 14LC
Prompt instructions like 'never do X' don't hold up in production. LLMs ignore them when context gets long or users push hard. Limits sits between your agent and the real world. Every action — database writes, API calls, refunds — gets intercepted and checked against your rules before it executes. Deterministically. No LLM involved in enforcement. Three modes: Conditions: hard rules on structured data Guideance: validate LLM output before it reaches the user and give the agent chance to reason and retry Guardrails: scan for PII, toxicity, prompt injection etc One line to integrate: npm…
Feb 2026 · limits.dev
- 15MA
This weekend I built a multi-agent coding system which, quite unexpectedly, beat Claude Code on Stanford's Terminal Bench! The architecture is straightforward, consisting of an orchestrator agent that deploys explorer & coder subagents to complete complex terminal based tasks, utilising an intelligent context sharing mechanism along the way which makes it all work. The repo has a lot of technical details, and all the code and prompts for you to play around with if you'd like! I had a lot of fun making this, I hope you have fun reading the README, using it yourself, or even extending it! As…
2025 · github.com
- 16IS
Hi HN, I built AgenTank. It is a small game where an AI agent writes the logic for your tank. You watch it fight, give strategic feedback, let the agent update the tank code, and send it back into battle. I have run 1,000+ battles on my own tank and spent about $200 in Claude credits improving it. The part I enjoy most is not just winning, but watching the tank make visible mistakes, thinking of a better strategy, and seeing whether Claude can turn that into better code.
May 2026 · agentank.ai
- 17WM
We wanted to test if a smaller model like GPT-4.1-mini could beat its bigger brother 4.1 at the game Tic-Tac-Toe using only context engineering. We put them in a 100-game tournament. For the smaller model, we gave it a few examples of winning moves from past games right before it made its own move. The results were clear. Without the examples, the smaller model struggled against GPT-4.1. With the examples, its effectiveness increased by nearly 200%, and it consistently won. It's a simple demonstration, but it shows that a smaller, faster model with good, timely examples can outperform a more…
2025 · github.com
- 18AC
Hello, This is something I've wanted to build for awhile (first bought the domain in 2020 (then lost it for a year+)). It's a tactics solver on each new tab page. There are 5 modes of play: Easy (<1200 elo) Medium (1200-1800) Hard (1800-2400) Expert (2400+) Speedrun (solve as many puzzles as possible in 60 seconds). I'm having a ton of fun with it, and definitely noticing improvements with actual games (definitely solved the most puzzles in my life while developing this extension). If you check it out, please let me know what you think. Any feedback is good!
Apr 2026 · leetchess.com
- 19WI
At Laminar (https://github.com/lmnr-ai/lmnr) we're building open source AI observability platform in Rust. We obsess over instrumentation DX for our Python and TS SDKs and in this new blog we outline how we made the most seamless way of instrumenting recently released claude agent sdk
Dec 2025 · laminar.sh
- 20WM
It is a prototype for Gemma 3n Impact Challenge hosted by DeepMind. We don't have experience on local LLM before, so it is a pretty fun learning experience. Hope to see more lightweight llm model in the future!
2025 · youtube.com
- 21LA
We combined Stanford's ACE (agents learning from execution feedback) with the Reflective Language Model pattern. Instead of reading traces in a single pass, an LLM writes and runs Python in a sandbox to programmatically explore them - finding cross-trace patterns that single-pass analysis misses. The framework achieved 2x consistency improvement on τ2-bench.
Mar 2026 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →