🔍 Introducing #MCP Server for Binary Ninja: Connect your AI assistants directly to Binary Ninja for powerful reverse engineering! Get pseudo code, analyze functions, rename symbols, and more! #reverseengineering#CyberSecurity#AI#LLMs#BinaryNinja
https://t.co/6ximun8Nwd
@kism3t_09@Kevin2600 Workbuddy gives any user 32G+256G serverless compute for free. And each project is a separate sandbox. For, free! 😋 At least for Chinese users, I think it's the same overseas
@dogmenei@AvePeptidePixie@UnrealizedPains@mchlhess Just think of it, who else will "date" such mass amount guys to develop her own "customary" 😅 for her "dating" is just word to persuade herself not be "prostituting"
@lyq_sqsp with E2EE its technically impossible to build an LLM relay. most relays i see (in our country) is built with vendored CPA + NewAPI to translate between anthropic messages <-> OAI response + inject system prompts to emulate Codex/CC/whatever clients.
easy to repro, deepseek flash got me a poc in ~30 min. but it's an arbitary file read with *tons* of "certain conditions". you can only read utf-8 decodeable files that contain `%` in their content.
@MilesCranmer You face all the limitations of a file system, but lose all its advantages.
You cannot use the find, ripgrep, ls, or sort commands to traverse a folder tree or read or edit specific portions of a file.
I don't think this makes sense. 🤔
This was called MCP Resources, long ago
@1techutopiapls@kunchenguid why would it ever benefit? Just a waste of tokens. Just use the advisor model for the execution. It won't lower your cost by combining models, because the advisor model is the bigger part of the bill either way
while maintaining my open source projects i noticed many people started using something called Oh My Pi
out of curiosity i took a look and gave it a go myself, and oh my..
it’s a giant pile of harness tricks bundled into one. each trick seems to do well on benchmarks. but…
this is EXACTLY what the bitter lesson told us to avoid. it may indeed work well at the time it’s evaluated, but every model release can invalidate a bunch of these results
unless every single part of the package gets rigorously re-evaluated on every model release, we simply can’t trust the bundle is actually helping. and in the long run, the bitter lesson has shown us over and over again that these tricks will not survive
that’s the core reason i’ve kept my own pi config extremely minimal. a simple harness gets better every time a stronger model comes out. a complex pile of tricks silently become a burden without you knowing
I got the latest iOS and macOS 27 booting in Qemu (with SPTM!)
- Virtual iPhone 17, 16, 15, 14, 13, 12 and every M1-M5 Mac supported
- Debug, patch, or modify everything: kernel, SPTM, TXM, launchd, dyld, user programs all modifiable/ GDB-able
- Boots directly to root shell in seconds
- Run your own programs as root in iOS/ macOS, no jailbreak / kernel patches required
- SPTM, TXM, MTE/MIE, genter/ gexit, GXF/SPRR/GL0-2, AMCC, AIC v1-3, Apple timer, many sysregs
- Automated setup; get running in just a few minutes
- Runs anywhere qemu runs... no ARM CPU required 😉
Try it here: https://t.co/PEydRXL25G