White hat hacker - Crypto/Defi. “It's always You vs You!” Book an audit/ formal verification at https://teapaw.org/t.co/xVRyRsluuy
Joined January 2023
- Tweets1.3K
- Following561
- Followers216
- Likes21.6K
copper screw retweeted
php devs had both mythos and daybreak red and they were still outrun by wiz calif using gpt-5.6-sol. you really can't buy the skill.
wait, PHP was actually scanned by Anthropic’s Claude Mythos earlier this year. they also had access to OpenAI’s Daybreak Red for validating exploits.
the PHP Foundation revealed this JUST yesterday.
keeping frontier cyber models in the hands of a selected few doesn’t seem to be working well for the internet.
thephp.foundation/blog/2026/…
Missed submitting something by 2 months, but insane to see that I could see the complete detail of the original report! This was not pay to submit too, QoL of bug submission is improving so much @immunefi 🔥
2
9
231
jumpcrypto.com/resources/har…
this is the codebase where AI found most / all bugs from openKritt iirc
Read to know what kind of workflows you have to add to find the bugs that were discovered here 🧵
1
1
96
HIGH: alternate angle ~
are there variables, that are different for each node (global_insert_cnt in this case)
that can affect anything related to consensus ? like tx ordering/ gas/ roots etc
"Map node-unique variables as entry points in each file and find consensus breaks"
1
16
copper screw retweeted
Good reminder that bug bounties can completely change your life in a matter of months.
One bug can pay more than a year’s salary, and one great year can change your financial future entirely.
From homeless to $1.4M in 11 months. Few careers offer upside this fast.
lol I figured out why they're doing this. Ig so many models improving so rapidly the last few months is because of distillation from Fable
By watermarking, they kinda pollute the training data of other models and when a model other than Anthropic tests positive for the watermark.. they know :)
this answers why its global instead of just EU too !
This might make me bullish for Claude
We’ve written an FAQ to answer some of the questions we've received about watermarking.
In summary:
• We’re implementing watermarking to comply with the EU AI Act. Other major model developers have signed the same Code of Practice and will also be implementing watermarking;
• Our watermarking method doesn’t have any practical impact on the quality or content of Claude’s outputs;
• The difference between watermarked and un-watermarked text will not be distinguishable to readers;
• Nothing is added to the text and there are no hidden characters;
• Watermarking doesn’t require extra tokens, and will not be more expensive;
• Watermarks can’t be traced to a specific person, organization, or chat.
Read more: anthropic.com/news/claude-te…
I’m out of the loop, can anyone tell me why the new royco competition has so many negative reactions ? 👀
1
130
COLDCARD ~ The code wanted hardware randomness but a logical bug with the macro use made it use a deterministic software randomness
AI didn't find it coz they didn't git submodule update the libs
claude.ai/share/b58f8e01-4dc… ~ has lines of code
This quoted post is unavailable.
1
174
copper screw retweeted
i would rather be an optimist and work hard than a pessimist posting about why things won't work.
it's much more difficult and the most likely path is failure, but society fails if people don't try.
no amount of "it will never work" essays will drive society forward.
copper screw retweeted
1/ How I use AI in smart contract audits (2026)
From 2023–2026, I entered 57 contests, reached top-3 30 times, won 21 and earned $627k in prizes.
109 highs + 156 mediums. 53 of them solos.
86 private audits.
This came from reading every line and building the mental model.
If you're into sorting Hackenproof programs by reputation, submission fee, KYC and choosing stuff, this will help you out 🤠🥳🥳
github.com/c0pp3rscr/Hackenp…
