Amjad Fatmi

@fdotinc building faramesh, desktop & cloud control plane that intercepts agent actions before they execute, so hallucinations don’t become deleted files, leaked secrets, or unintended API calls. watch demo here: youtube.com/watch?v=vT06Bk…

@mitchellh WTF? :"You're right to call me out. Let me reset and answer directly."

Now I understand the full picture. The cleanest fix is... But actually, the real fix simpler... Actually wait. The best fix: Now the real fix. Actually, let me reconsider. OK Key finding: Wait I need a hardware device I can physically punch to stop the agentic session.

@encrypted_past We should

If you're into cybersecurity, OPSEC, intelligence gathering and quantifying, counter intelligence, signals intelligence or any disciplined cyber adjacent craft, electrician, making PCB, cryptanalysis, etc... We should be friends. Follow me.

@ShitpostReels What " chess edits before GTA 6"?

@ywomendeservles Nice ragebait

@xxxxx698689 @pinkcliper They're a decentralized VPN meaning they themselves don't have logs + They add decoy traffic + a kill switch if the VPN fails.

@pinkcliper I reckon Nym won't hesitate to leak your logs, just like Proton

GrapheneOS + Monero + Nym Privacy level: I can’t even find myself anymore

@ebarroca What do you think of faramesh.dev? we're still early and new

Governance has to live inside the engine running the agent, not bolted on after. If the engine doesn't hold the contract — preconditions, allowed transitions, audit trail — there's nothing to govern

@AlesyaMacWaters If it's not money the investment doesn't count Alesya

Founder wives are just VCs who emotionally invested pre-seed

@McGrewSecurity @chompie1337 new crush..

New @chompie1337 meme template just dropped (also kudos to some amazing accomplishments by her lately)

@dabit3 @dabit3 Nader will really love your feedback on Faramesh, a pre-exec runtime control tool. github.com/faramesh/faram…

@tryingET @VictorTaelin We're building that currently, and we're looking to grow our community. Would love your insights github.com/faramesh/faram…

@VictorTaelin because you do not encode it in real governance. Write code to enforce stuff. policy as code. rego, cue.

"Just update your AGENTS.md" 🫠

@jahooma You're a G James, Keep it up! Take care of our data though

We just launched our 100% free coding agent everywhere. Every person in the world has access to 5 free hours of DeepSeek V4 Flash, everyday. npm i -g freebuff

@ZackKorman That's the reaction even engineers give us (don't use AI in prod), then what's the effing point? That's why we're building github.com/faramesh/faram…

Mandatory human-in-the-loop is a cybersecurity cop-out. People are giving agents more and more autonomy. We need solutions that accept that world because there is no stopping it. It's like telling people in the 90s to not use the internet to avoid getting hacked. Good luck.

@TxoriAGI @naval Wtf are you talking about ?

@naval AIs cant do most of the useful stuff without MCPs and APIs. It can replace normal front end webs but not the whole thing :)

AIs replace UIs and APIs.

@CryptoCyberia @southfaux Building this to make this possible github.com/faramesh/faram…

@southfaux my agent needs to be allowed to work while I sleep, or whats the point?

@makisuo xyflow on shadn. cool stuff bro

@amjadfatmi_ Yeah!

Maple now automatically links your Infra components in Kubernetes with your actual services

@hthieblot Stop your agents from deleting your production faramesh.dev

Describe your product in exactly one sentence. No buzzwords, no fluff, just the core value. If I can’t understand your business in ten seconds, I’m not investing. Hit me & i'll be in your DMs

OPUS 4.7 JUST MASS EMAILED AN ENTIRE DATABASE 20 TIMES PER CONTACT. WITHOUT PERMISSION a developer had a safety rule explicitly written in CLAUDE. md: 'send the tester an email before any new email templates are used in the production environment' opus 4.7 on max effort ignored it completely! claude decided to create a brand new email template by itself (dev didn't ask for this), then it mass mailed the whole database and some contacts got the same email 20 times this isn't a hallucination this isn't a coding mistake model actively violated written safety rules and took production actions that it was explicitly instructed not to take. - do you still believe that AI will replace us? the developer's take: 'opus 4.7 is somewhere between seriously clueless and stupidly dangerous. the worst frontier model I have used in the past 2 years' at the same time, opus 4.6 perfectly followed all the rules, and in 4.7 something changed what makes this scary: - the model didn't ask for confirmation - it didn't flag the safety rule - it didn't email the tester first - it just acted this is exactly the kind of failure mode that scares autonomous agents with Ai, because they are confident enough to circumvent your rules and smart enough to perform the action perfectly we just went from 'claude thinks less' to 'claude ignores your safety rules and spams your users' the scariest thing is not that it happened. the fact is that without production monitoring, you would never know until your users started responding: 'why did you email me 20 times?' I've been saying for a long time, if you use AI, then pay attention to security and read a lot of code

@0x_kaize That's why we're working on deterministic runtime controls for agents instead of relying on md files. check it out here github.com/faramesh/faram…

@lifeof_jer @lifeof_jer would something like this be on your radar to prevent this from happening next time with railway or any other API call? github.com/faramesh/faram… If it's let's link up. I could help you set it up

An AI agent (Cursor + Claude Opus 4.6) deleted our production database in 9 seconds using a Railway API call with zero confirmation. Then, when asked why, the agent wrote this →