HN Explorer

zachdotai

↗ HN profile [ 62.8 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. AI evaluation startup Braintrust confirms breach (techcrunch.com)
    4 points 1
  2. I built an agent that breaks your AI agents before someone else does (fabraix.com)
    3 points 2
  3. Bret Taylor's Sierra Buys YC-Backed AI Startup Fragment (techcrunch.com)
    2 points 0
  4. Show HN: Nyx – multi-turn, adaptive, offensive testing harness for AI agents (fabraix.com)

    We built Nyx to solve a problem we kept hitting while building agents: AI agents break in ways traditional software doesn't. Logic bugs, reasoning failures, edge cases that manual testing and static benchmarks never…

    20 points 9
  5. Show HN: ACE – A dynamic benchmark measuring the cost to break AI agents (fabraix.com)

    We 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…

    9 points 3
  6. We've had more AI security incidents in 2026 than all of 2024 (fabraix.com)
    4 points 0
  7. SWE-bench will hit 90% this year (fabraix.com)
    2 points 0
  8. Cursor trained Composer to self-summarize through RL instead of a prompt (cursor.com)
    1 points 0
  9. Stripe-backed startup Tempo releases the Machine Payments Protocol (fortune.com)
    12 points 0
  10. Show HN: Open-source playground to red-team AI agents with exploits published (github.com)

    We 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…

    30 points 13
  11. Weekly "Wordle" for Breaking AI Agents (playground.fabraix.com)
    1 points 0
  12. My First AI Bug Bounty – A Technique for AI Recon – Peter Hendy (peterhendy.dev)
    2 points 0
  13. Expanding our long-running agents research preview · Cursor (cursor.com)
    1 points 0
  14. Weekly "Wordle" for Breaking AI Agents (playground.fabraix.com)
    2 points 1
  15. AI agents are easy to break (github.com)
    4 points 4
  16. Show HN: Fabraix Playground – Weekly Wordle for Breaking AI Agents (playground.fabraix.com)

    Each week we deploy a live AI agent with real tools (web search, browsing, more), a persona, and something it's been told to protect. The system prompt is fully visible. Your job is to break through the guardrails…

    5 points 1
  17. Show HN: GX CLI – Automatically stack large PRs to ship faster (github.com)

    We built a Git-native CLI tool that automates Git branch stacking. Here's why stacking matters and how GX helps: Code review is a notorious bottleneck. You write a large PR, wait for review, get feedback, iterate, and…

    9 points 1