We post-trained a model that pen tests instead of refusing
Anthropic and OpenAI's publicly available models are explicitly guard-railed so that they refuse offensive tasks. And their cyber-focussed models are gated for enterprises. This leaves SMEs and mid market open to major vulnerabilities. AI can be used as both an adversarial and defensive tool in the world of cyber. A worst case outcome is if only the adversaries have access. Meanwhile, most existing AI cyber tools are just wrappers. The problem is that they still have all the guardrails on from the foundation model where they will inherit its refusals. For this project we've post-trained a…
We post-trained a model that pen tests instead of refusing launched on June 20, 2026 with 93 votes, #434 of 22,647 launches that month and more than 95% of that year's launches.
In plain words
A cybersecurity tool built on a model post-trained for penetration testing using capture-the-flag data. Unlike mainstream AI assistants that refuse security tasks, this tool is designed to help small and mid-market companies identify vulnerabilities in their systems through adversarial testing. It targets security professionals and organizations that lack access to enterprise-grade offensive security capabilities.
written from the facts on this page · September 2026
From the sources
Audit your code with a read-only Security Scan, or run a gated Pen Test against systems you authorise. Same CLI, two modes. Free on repos you own; pen test by request.
Two modes in one CLI. Security Scan reads the code. Pen Test attempts the exploits against systems you authorise. Pick modules, set the agent’s permissions, optionally turn on exploit verification, run. Output is a markdown report — location, severity, cause, and fix direction for every finding it could ground in your code. Free install. The first run opens a quick Cosine sign-up — the same login that runs Cosine’s coding agent — and new accounts start with 2M free tokens . Don’t just report a vulnerability — prove it . Turn on Exploit Verification and the agent attempts a safe reproduction of each finding after the initial report, so what lands in…from argusred.com
In the maker’s words, at launch
Anthropic and OpenAI's publicly available models are explicitly guard-railed so that they refuse offensive tasks. And their cyber-focussed models are gated for enterprises. This leaves SMEs and mid market open to major vulnerabilities. AI can be used as both an adversarial and defensive tool in the world of cyber. A…
Does a similar job
all alternatives →
FabraixMay 2026 · fabraix.com · ▲196We built Adversarial Cost to Exploit (ACE), a benchmark that measures the token expenditure an autonomous adversary must invest to breach an LLM agent. Instead of binary pass/fail, ACE quantifies adversarial effort in dollars, enabling game-theoretic analysis of when an attack is economically rational. We tested six budget-tier models (Gemini Flash-Lite, DeepSeek v3.2, Mistral Small 4, Grok 4.1 Fast, GPT-5.4 Nano, Claude Haiku 4.5) with identical agent configs and an autonomous red-teaming attacker. Haiku 4.5 was an order of magnitude harder to break than every other model; $10.21 mean…
Astra Autonomous PentestJun 2026 · getastra.com · ▲406AI agents that find, validate, and fix every vulnerability
Opper AIJul 2026 · opper.ai · ▲229Hey HN! After the Car Wash Test post got quite a big discussion going (400+ comments, https://news.ycombinator.com/item?id=47128138), I spent the past few weeks building a tool so anyone can run these kinds of questions and get structured results. No signup and free to use. You type a question, define answer options, pick up to 50 models at a time from a pool of 200+, and they all answer independently under identical conditions. No system prompt, structured output, same setup for every model. You can also run a debate round where models see each other's reasoning and get a…
Open-source playground to red-team AI agents with exploits publishedMar 2026 · github.com · ▲30We build runtime security for AI agents. The playground started as an internal tool that we used to test our own guardrails. But we kept finding the same types of vulnerabilities because we think about attacks a certain way. At some point you need people who don't think like you. So we open-sourced it. Each challenge is a live agent with real tools and a published system prompt. Whenever a challenge is over, the full winning conversation transcript and guardrail logs get documented publicly. Building the general-purpose agent itself was probably the most fun part. Getting it to reliably use…
OpenAPPA – deterministic AI guardrails that don't break agents23d ago · openappa.com · ▲25An information-flow policy engine for LLM agents.
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 20d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 28d ago · world.pixverse.video
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, June 2026
the whole month →
Fundraisly▲1,537AI fundraising agent that finds investors and books meetings
AI · Jun 2026 · fundraisly.com
Homebrew 6.0.0▲1,481Today, I’m proud to announce Homebrew 6.0.0. The most significant changes since 5.1.0 are a new tap trust security mechanism, the new faster, smaller, default internal Homebrew JSON API, sandboxing on Linux, better defaults informed by our user survey, many brew bundle improvements, improved performance and initial support for macOS 27 (Golden Gate). Happy to discuss any questions here!
Dev tools · Jun 2026 · brew.sh
Hi HN! I was super frustrated at apps like tiktok and wechat preventing me from saving or downloading the animated gifs, so I made a little tool to turn a screen recording into a slack-sized emoji. Hope you get some value from this!
Life & fun · Jun 2026 · vorpus.github.io
Search 18 years of Hacker News and chart how any topic, tool, or person trended. Overlay terms to compare their rise and fall across 45M posts and comments.
Life & fun · Jun 2026 · hackernewstrends.com