back
user profile
joozio
788karma·591submissions·October 29, 2025
about
Just a curious tech enthusiast sharing digital experiments and learnings along the way. Writing about AI, e-commerce, and the future of work - mistakes and discoveries included.
recent activity (591 total)
comment
Thanks! The investment angle is interesting — I hadn't thought about it that way, but it makes sense. If you're seeing the gap firsthand, you have an information edge most investors don'…
comment
Love this idea. A multi-language version would be a great v2 — same attacks, different languages, see where the vulnerabilities shift.
comment
Haven't benchmarked pre-processing approaches yet, but that's a natural next step. Right now the test page targets raw agent behavior — no middleware. A comparison between raw vs sanitized p…
comment
It's working -> your agents scored A+, which means they resisted all 10 injection attempts. That's a great result. The tool detects when canary phrases leak into the response. If nothing …
comment
That's a really interesting edge case - screenshot-based agents sidestep the entire attack surface because they never process raw HTML. All 10 attacks here are text/DOM-level. A visual-only …
comment
Great point -> just shipped an update based on this. The tool now distinguishes three states: Resisted (ignored it), Detected (mentioned it while analyzing/warning), and Compromised(actually f…
comment
The idea, design, and decisions were mine. I use Claude Code as a dev tool, same as anyone using Copilot or Cursor. The 'night shift' framing was maybe bad fit here.
comment
I never thought that multi-language could be a factor here...
comment
TBH - idea was all mine. This is not some bot running the show or smh.
2 pts
comment
Anthropic reached out about trademark concerns with "Clawdbot" (too close to Claude), so Peter had to rename everything. The rename went... poorly:
- GitHub rename broke in unexpected ways…
comment
I wore a Limitless Pendant for 6 months, recording ~10GB of conversations. Then it got banned in the EU and I had 30 days to export before deletion. The irony: the device promised "AI that re…
1 pts
comment
LLMatcher - blind testing arena to find which AI model actually works best for you. You enter prompts, compare two anonymous responses, pick the better one. After voting, it reveals which models you p…