guarism∅

@StefanMaier @ollama Because everyone is trying to run Kimi k2.6

oh wow, GLM 5.1 on @ollama is so much faster today! wow!

@openclaw Gemini video and music generation are still broken

OpenClaw 2026.4.20 🦞 🧠 Kimi K2.6 support + provider-aware /think 💬 BlueBubbles iMessage sends + tapbacks fixed ⏰ Cron state/delivery cleanup 🔐 Gateway pairing + plugin startup hardening Less haunted. More useful. github.com/openclaw/openc…

@hetmehtaa is MCP one word?

i am a cybersecurity guy, scare me with one word

The Internet of Stuff and Things: Diagnosing PRTG UDP Errors on a Syslog Server with LLMs blog.guarismo.net/2026/04/diagno…

🛑 A design flaw in Anthropic’s MCP allows remote command execution on AI systems. 150M+ downloads affected as unsafe STDIO defaults expose 7,000+ services, including tools like LangChain and Flowise. Anthropic calls the behavior “expected,” leaving the risk across the AI supply chain. 🔗 Read → thehackernews.com/2026/04/anthro…

@songjunkr Then choose one of those projects and start contributing to it

Whatever you're trying to make with vibecoding: 99% chance it's already open-sourced on GitHub.

@TechOperator DIR /W/S

You walk up to your computer and see this screen. Your next move?

I pointed claude opus at chrome and told it to build a full v8 exploit for discord. A week of back-and-forth pulling it out of dead ends. 2.3B tokens. $2,283 in API costs, and it popped a shell. hacktron.ai/blog/i-let-cla…

@Scobleizer @openclaw @cvander @frontiertower Hey @RayFernando1337, have you seen this? Sorry if you already talked about this I missed it.

So San Francisco. An @openclaw run vending machine. Met @cvander who built it. At @frontiertower which is a high rise in San Francisco stuffed with startups.

@Scobleizer @openclaw @cvander @frontiertower Wow! what's the cost of generating the ads and all those images with LLMs?

The Internet of Stuff and Things: Analyzing a Mirai Botnet PCAP with Claude Code blog.guarismo.net/2026/04/analyz…

Marcus Hutchins, the guy famous for stopping the WannaCry Ransomware, probably has the best take on Mythos doing vulnerability research

@Anaya_sharma876 Change the wallpaper

I’m a Windows user, just installed Linux for the first time What should I do first?

for folks who feel their Claude Code got nerfed, here's what I've set in my ~/.claude/settings.json to make my CC's behavior more stable. snippet and explanation in thread below -

A fully local 26B MoE model was built for red teaming and bug hunting. Trained on elite bug reports and real evasion tactics. DPO fine-tuned for hunter mindset. Claude sees your payloads in logs; that's why BugTraceAI Apex 26B local MoE for real red teamers. - executes WAF bypasses with internal thinking blocks. - It enforces deep internal reasoning before generating any output. - Delivers production-grade WAF/EDR evasion with AES-256-GCM obfuscation. - Zero refusals, Trained on real-world elite reports and evasion techniques. Fits in 16.7GB. Runs on RTX 3060. - huggingface.co/BugTraceAI/Bug…

@AdolfoUsier @nvidia @opencrabs Isn't it throttle to 40 rpm? I tried them for a while but it was very slow and kept getting errors.

Again no one is talking about this but @nvidia is offering a bunch of completely free API endpoints. They just dropped MiniMax M2.7 free. I mean literally, completely FREE!!!! And you can easily setup it as custom provider on @opencrabs Grab your key here: build.nvidia.com/minimaxai/mini… On config.toml paste this or just give to @opencrabs opencraw, hermes, or any harness to setup for you: [providers.custom.nvidia-minimax27] base_url = "integrate.api.nvidia.com/v1" context_window = 200000 default_model = "minimaxai/minimax-m2.7" enabled = false/true On keys.toml paste: [providers.custom.nvidia-minimax27] api_key = "your_nvidia_api_key" And boom you are all set to go, at 0 costs! Kimi is there since weight release so think about it. I'm still handling all complex tasks to Qwen 3.6 Plus but for fallback I have just setup Nvidia with MiniMax M2.7, why pay if you have it free? And also GLM 5 btw xD Can't wait for 5.1 and Qwen 3.6. Meanwhile enjoy it!

@code_rams I was trying this exact thing a few days ago on @discord only to find out bots can't talk to each other there.

I run two AI agents separately. One - Claude Code connected to Telegram via a custom bot. One - OpenClaw running on the same VPS with Codex subscription. Today I had a random idea - added both to a Telegram group and asked them to introduce themselves. Both did it. One even said "Think of me as the always-on ops sibling." 😂 Watching two AI agents greet each other is something else. Now I have no idea what to do next. Currently researching how to use both effectively - different models, different strengths, same owner. If you've done something like this, what tasks would you split between them?

@Alex_Rogov_js @dramaricic It's better to add the testing part to the /plan when promoting for changes if this is an online change. If this is unattended self improvement then CLAUDE.md should be good

@dramaricic the fix: put "never claim a task is complete without running the tests" in your CLAUDE.md. our first-try success rate went from ~60% to 82% with that single rule. the model will confidently lie about fixing things if you don't force verification

Claude: I fixed all files, they are bug free. Me: Log in button doesn't work. Claude: You're right! Let me check files....Checking... You've hit your limit...

🚨 Diosdado Cabello is one of the regime’s most powerful enforcers—linked to corruption, repression, and criminal networks that have destabilized an entire region. The interim dictator, Delcy Rodríguez, should immediately surrender him to U.S. authorities. Accountability cannot be selective. Justice must be complete. Time to finish the job. #Venezuela

@realsigridjin Sounds like a /plan mode in steroids

a friend from the first ralphthon just shipped something thats been living in my head rent free for weeks its called ouroboros. does a socratic interview on you before it lets you write a single line of code. refuses to let you skip the part where you actually know what you want then he ran it on itself for 7 generations. it spat out almost 13k lines of code and the fields kept oscillating 4 9 8 7 8 in circles like a snake eating its own tail he fixed the entire plato cave problem with two lines of prompt drop your vaguest project idea below and i'll run ouroboros on it tonight genuinely curious what assumptions it surfaces that you didnt know you had