lightcap.eth 🏴 (@lightcap) profile picture

Heretic @ Codezero. (also CTO/CPO)

Joined January 2007
Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)Photo by lightcap.eth 🏴 (@lightcap)
Credentials continue to be the target of just about every attack in 2026.
🚨 The "π™ΌπšŽπšπšŠπš•πš˜πšπš˜πš—" Campaign is live... 𝟻,𝟽𝟷𝟾 malicious commits to 𝟻,𝟻𝟼𝟷 GitHub repositories in a six-hour window. Using throwaway accounts and forged author identities (build-bot, auto-ci, ci-bot, pipeline-bot), the attacker injected π™Άπš’πšπ™·πšžπš‹ π™°πšŒπšπš’πš˜πš—πšœ workflows containing πš‹πšŠπšœπšŽπŸΌπŸΊ-πšŽπš—πšŒπš˜πšπšŽπš bash payloads that exfiltrate: - CI secrets, - cloud credentials - SSH keys - OIDC tokens - source code secrets Check your repo / Technical details: safedep.io/megalodon-mass-gi…
1
3
199
Deepseek ban. Queue biohazard. Activate phase 3.
36
Privacy is the biggest problem with personal agents. You're not only compromising your own privacy, you're complicit in destroying the privacy of your loved ones and friends without their consent too.
36
Shippers gonna ship.
267 updates in 30 days. Roughly nine a day, every day, all of September. Here's what shipped 🧡
2
52
Why would they have a wet lab in the first place ffs
Two Anthropic employees said this afternoon that unidentified biological material had escaped containment at the company’s Bay Area wet lab. The employees described the material as β€œliving organisms.” Its nature and the extent of the breach remain unclear.
48
I grow more convinced every day that the frontier lab’s fear mongering is just a prelude to the next chapter. Cyber didn’t scare us sufficiently so bio is next. We’ll go from agents escaped the cyber sandbox to biological agents escaping into meatspace any day now.
31
This is only going to get worse
Woke up to all my data leaked by @Revolut. Sharp reminder that KYC hasn’t produced meaningful upside and has put many in harm’s way.
33
Please can our supreme AI overlord not be called Doug.
1
45
How is it that now misspellings make me trust the writing more now? Didn’t see that coming.
1
53
Meet Hellboy. My new Kamado Joe Jr.
1
71
Could be wrong, but Opus-4.8 is feeling particularly nerfed today. It's been substantively wrong on almost every task at high effort.
2
81
Why oh why did we have to bring waffles to the table?
1
54
VAR is killing soccer.
1
1
31
The trend of allegedly smart people quote tweeting with one word is killing me. It’s intellectual laziness at its worst. β€œInteresting”
2
27
Just retweet if you’re gonna mail it in so hard.
6
It’s a great twist that the jobs historically most under appreciated and under paid in our society are the exact jobs that are the hardest to replace with AI. Caretakers have been systemically marginalized for the last century and here we are today admitting they’re irreplaceable
13
Did the security industry push for monocultures within orgs only to expose them to the potential of one threat that wipes them out? Same patch, same OSes, same apps. NSA/Mythos agricultural analog? Bananas 🍌
1
2
58
For anyone else that's having rendering issues with garbled text using Claude code with ghostty or tmux a mildly tolerable workaround is hitting ctrl+l (lower L) for a rerender.
1
59
Custom status lines also seem to be a potential culprit.
22
Giving up on claude code TUI for a bit. The rendering bugs make it literally impossible to get meaningful work done. Resizing the terminal window helps for the next turn's prompt but then goes right back to corrupted output.
48
One thing I've noticed today is that Fable is *much* better with skills loading and usage. I rarely have to remind it to use an intended skill, it's been picking them all up just as part of the conversations. Opus needed constant reminders. This is a huge improvement to me.
29