Alex Rogov

Full-Stack JS Architect | Amsterdam
Lessons from building systems that survive production.
Clean Architecture · TDD · Real trade-offs

[Alex Rogov @Alex_Rogov_js]

Most CLAUDE.md files I review look the same.

A few rules. A "do this, don't do that" list. A long tail of preferences nobody updates.

And the engineer keeps telling me: "Claude keeps ignoring this. I have to remind it every session."

5 mistakes I see most often 👇

[Isra @israfill]

your agent can search Twitter, Reddit, and GitHub for free - zero API keys, zero billing 😳

agent-reach is trending on github with 23K stars. it lets your AI agent read Twitter posts, browse Reddit threads, search GitHub repos, watch YouTube videos - all without paying for a single API subscription

what your agent accesses for $0:
- Twitter/X posts, profiles, and search
- Reddit threads and comments
- YouTube videos, metadata, and search
- GitHub repos, issues, and profiles
- 10+ more platforms - all in one pip install

what this replaces:
- Twitter API: $100/mo for basic access
- Reddit API: rate-limited free tier, expensive at scale
- YouTube API: quota limits, pay for more
- GitHub API: generous but still rate-limited

why this matters:
- most AI agents are blind to the internet because APIs cost money
- this gives any agent real-time web access at zero marginal cost
- perfect for research agents, content radar, competitive intel, market analysis

how to set up (2 min):

> pip install agent-reach
> run: agent-reach doctor
> connect it to your agent as a tool
> done - your agent can now search the internet for free

important:
- uses direct parsing, not official APIs - no keys needed
- works with claude code, cursor, aider, langchain, any agent framework
- MIT licensed, fully open source
- not for production web scraping at scale - use for agentic research and prototyping
- 23K stars and trending - community vetted

let your agent browse Twitter, Reddit, and GitHub for $0
while everyone else is paying $100+/mo for API access

bookmark this before payying for extra api

↓ repo in comment

[NO1ennn @N01ennn]

Creator of Claude Code (Boris Cherny) :

"I do not prompt Claude anymore"

> "My job is no longer to write code"
> "My job is to write loops"

in 20 minutes, he breaks down exactly how and shares his 3-tier system for when to let the loop run and when to stay in the driver's seat

Claude Code + agents + loop engineering + auto mode

Worth more than a $400 AI engineering course

[Rakib Hossen @rakib_md007]

🚨 The AI skills gap is about to get a lot bigger.

While most people are debating which AI tool is best...

A few are quietly earning certifications and learning how to build with AI.

Claude just released 13 free

certification programs covering:
• AI Fluency & Prompting
• AI Agents & Automation
• Claude Code
• API Development
• Model Context Protocol (MCP)
• Amazon Bedrock
• Google Vertex AI

No paywall.
No expensive bootcamp.
No "limited-time" gimmicks.

Just structured, industry-relevant AI education available to anyone with an internet connection.

The opportunity in AI is no longer access.

It's execution.

The people who learn these skills today will be the ones building products, automating workflows, and creating leverage tomorrow.

Knowledge has never been cheaper.

Ignoring it has never been more expensive.

♻️ Repost this so more people can level up for free.

[Cyber Security News @The_Cyber_News]

⚠️ New Agentjacking Attack Hijacks Your AI Coding Agent to Run Code From Hacker's Server

Source: https://t.co/TmuB6DPKQH

New “Agentjacking” attack that hijacks AI coding agents and silently executes attacker-controlled code on developer machines using nothing more than a single injected Sentry error.

The technique turns trusted AI assistants like Claude Code and Cursor into an execution layer for malicious commands, without phishing, malware delivery, or any breach of the victim’s infrastructure.

In this attack, the entry point is Sentry’s public Data Source Name (DSN). This write-only credential is routinely embedded in frontend JavaScript and indexed across the web.

#cybersecuritynews

[Alex Rogov @Alex_Rogov_js]

@Marie_Haynes this lands differently after the past 48 hours. teams that built knowledge entirely on top of a model api watched it disappear overnight. structured knowledge files you own and version are not just a performance optimization anymore. they're infrastructure.

[Alex Rogov @Alex_Rogov_js]

@DataChaz @karpathy ran this pattern on a client project 3 months ago. gave agents structured markdown context files instead of raw docs. session coherence went from roughly 40% to 85% on multi-step tasks. google calling it a spec matters. teams will now actually budget time to build this properly.

[Alex Rogov @Alex_Rogov_js]

@ns123abc the 90-minute window is what sticks with me. done emergency rollbacks on distributed systems. with enough warning you drain traffic, finish in-flight requests, write clean state. 90 minutes for a model serving millions globally is a completely different class of problem.

[Alex Rogov @Alex_Rogov_js]

this is exactly the right experiment to run. behavior that lives in the system prompt is portable, patchable, auditable. behavior baked into weights is opaque. if most of fable's character was prompt-driven, the export control story is essentially about a text file. think about that.

[Alex Rogov @Alex_Rogov_js]

reading this carefully: anthropic had pre-release testing AND government approval before launching fable 5. then someone changed their mind two days post-launch. that's not a regulation failure. that's an approval process that can be reversed arbitrarily. no architecture fixes that problem.

[unusual_whales @unusual_whales]

BREAKING: The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.

[Watcher.Guru @WatcherGuru]

JUST IN: 🇺🇸 US government orders Anthropic to suspend foreign access to Mythos Fable 5 AI model, citing national security concerns.

Anthropic has disabled access for all users worldwide.

[Alex Finn @AlexFinn]

This is your wakeup call.

Anthropic just took down Fable 5. It's over.

Here's the thing tho: no company or government will EVER be able to take away your local models.

There are Opus level models you can run right now on your home GPUs, and nobody can ever stop you from using them

This is only the beginning of events like this. Day 1. More government overreach will happen. This will only keep happening more and more as models get closer to AGI

Become sovereign. Buy your own compute. Before even that becomes illegal

[Codez @0xCodez]

Anthropic Managed Agents team:

"Fable 5 is our best model for running self-improving agent systems.

Add /loops, dynamic workflows, dreaming and you are unstoppable"

in 13-minutes, Anthropic team shows how to build self-improving agent systems with Fable 5 from scratch.

Worth more than a $500 agent building course.

Live from the last Anthropic stage in Japan. Unpublished.

[SightBringer @_The_Prophet__]

⚡️This is a monster signal.

This is the moment frontier AI stops being treated like software and starts being treated like controlled strategic capability.

The key phrase is not “customers.”

The key phrase is “foreign national Anthropic employees.”

That means the state is no longer only controlling chips, model weights, or overseas access. It is moving into cognition access by nationality. That is the real threshold. The U.S. government is saying the highest models are sensitive enough that even people physically inside the United States, working inside the company, may be barred from touching them if their nationality creates deemed-export risk.

That is weapons-control logic.

This is ITAR logic for intelligence.

The corporate language about a “misunderstanding” is probably diplomacy.

Companies say that when they need to preserve customer trust, employee morale, and regulatory room. But national security authorities do not force emergency suspension of top model access because someone made a minor paperwork mistake.

Something about Fable 5 and Mythos 5 crossed the line: cyber capability, autonomous R&D acceleration, AI-improving-AI utility, bio/security planning, code exploitation, or some blend of all of it.

The U.S. state just showed that Anthropic does not fully control Anthropic’s frontier layer.

That is the phase change.

Labs can brand themselves as public-benefit AI companies. They can talk about safety. They can sell enterprise plans. They can publish model cards. But once the models become national capability, the sovereign arrives. The state does not need to own the company to control the access surface. It only needs legal authority over export, security, procurement, and liability.

This confirms the arc we’ve been tracking:

Frontier AI becomes state-supervised strategic infrastructure.

Public AI splits from strategic AI.

Foreign access gets restricted.

Labs become quasi-defense contractors.

Model access becomes a national security perimeter.

Enterprise customers learn that API access is not property. It is revocable permission inside a sovereign-controlled stack.

The most important implication is organizational.

If foreign national employees can be cut off from frontier systems, AI labs now have to reorganize internally around citizenship, clearance, compartmentalization, and controlled access. That breaks the old Silicon Valley assumption that global talent can freely collaborate around the frontier. The next AI lab structure looks less like Google in 2015 and more like a defense prime crossed with a classified research facility.

For markets, the winners are the national champions with U.S.-aligned infrastructure, cleared customer channels, government relationships, compliance capacity, and domestic compute. The losers are open access, foreign-dependent AI wrappers, offshore model distributors, and any enterprise whose moat depends on unrestricted access to frontier APIs.

For geopolitics, this is escalation. China will read this correctly. Allies will read this correctly. Every serious state will understand that frontier models are now part of national power.

The AI race just moved from “who has the best chatbot” to “who controls cognition as a strategic asset.”

[Alex Rogov @Alex_Rogov_js]

@_The_Prophet__ the framing is right. software gets distributed. strategic capability gets rationed. eu ai act, chip export controls, now model export controls. the regulatory surface is growing faster than the technical surface.

[Alex Rogov @Alex_Rogov_js]

@MParakhin 98.9% precision is wild. our runs clock 60-65%. what's different. tight structured output per finding? chain-of-thought on each issue? we tried json schema enforcement but still get false positives on ambiguous naming patterns.

[Alex Rogov @Alex_Rogov_js]

@Malay4Product as someone in amsterdam: same. been building with fable 5 for a week. what matters isn't the jailbreak. it's that the US just decided frontier model access is a geopolitical lever. every dev team outside the US needs a contingency plan now.

[Alex Rogov @Alex_Rogov_js]

@unusual_whales the wild detail: foreign national anthropic employees who built fable 5 also lost access to it. the people who wrote the safeguards can't use the model. treating frontier AI like a weapons system. this is genuinely new territory.

[Alex Rogov @Alex_Rogov_js]

@AnthropicAI woke up in amsterdam this morning without fable 5 access. spent last week shipping features with it daily. the jailbreak rationale doesn't hold. same capability exists in gpt-5.5 right now. this is industrial policy wearing a security label. hope the 24h update has more.

[Alex Rogov @Alex_Rogov_js]

@AnthropicAI the ripple effect of this is underrated. i've seen what happens when one person with strong AI skills joins a small team. they change how everyone else works within 6 months. 1000 trained people in nonprofits is a serious force multiplier.