Pinned Tweet
This Agent Harness Made Kimi K3 Dangerous
@Kimi_Moonshot @pidotdev @opencode @NousResearch @OpenAI @claudeai
#kimik3 #pi #opencode #fable5 #sol #claudecode #codex #ai #harness
youtu.be/G05evc0wumc
1
3
322
Build a company ontology without graph databases or proprietary models.
Just open-weight models and Postgres.
Open-source. No Anthropic and OpenAI
91% accuracy on LoCoMo and 79% on BEAM.
Agents work across increasingly large amounts of data, but don't understand how information relates and spend millions of tokens just reading files.
Cognee now connects GitHub, Slack, Linear, Notion, and 10+ other integrations with your agents or software factories and automatically builds an ontology of your company
It's:
- Free
- Runs on Postgres
- Uses small models to construct different memory elements
- Uses retrofitted embeddings to avoid graph databases
- Automatically builds an ontology
Here's how it works:
Taylor Lindores-Reeves retweeted
Today we’re launching Shield: AI-native defense against offensive agents.
Attacker agents probing your site? Don’t block them. Deceive them.
We pointed ARTEX — the autonomous AI pentester the FBI just warned about—at our own prod. 5 minutes in, it pulled /.env and /.git/config (200 OK) with credentials inside.
Shield turned ARTEX attack surface into a synthetic maze with fake endpoints, decoy credentials, and poisoned responses that hijacked its agent memory.👇
Taylor Lindores-Reeves retweeted
show this video to anybody that is an old school programmer that hates AI
this is one of the best videos I’ve seen all month.
Taylor Lindores-Reeves retweeted
pastebin.com/PihVyZur
7.3 TB ShinyHunters Databases Leak
Use qBittorrent
Taylor Lindores-Reeves retweeted
stop making boring PDFs
pdfcn.dev/
Introducing Claude Meetings: add Claude to your team’s video call and it builds docs, whiteboards, and other artifacts with you while you’re still talking.
Starting to roll out in beta on Team and Enterprise plans.
Taylor Lindores-Reeves retweeted
Self-hosted Claude Managed Agents is now in private beta.
You can host the agent harness, sandboxes, and session history, in your own cloud or on-premises, and choose where requests to Claude are sent.
Taylor Lindores-Reeves retweeted
thank me later, Apple users
meetribbon.app
thank me later, Apple users
fileosmac.com
Taylor Lindores-Reeves retweeted
If something isn't working, anything. Don't ask yourself why it isn't. Ask yourself how could you fix it. There is nothing you can't fix, if you only think. Think! Think, don't give up. There is always something you can do. There always is another option
Taylor Lindores-Reeves retweeted
If you haven’t, please read about what Toyota did in manufacturing the mid 20th century.
they solved a bunch of the problems people are having managing agent swarms. And they wrote it down as generalisable principles
Now in beta: Businesses in ChatGPT.
Claim your business, manage your information, and get insights on how people engage with you.
Eligible businesses can also build a Business Agent to answer customers’ questions and help them explore what you offer.
openai.com/index/introducing…
Taylor Lindores-Reeves retweeted
please keep this secret
layoffsandhires.com
Taylor Lindores-Reeves retweeted
This is the London Banana. As long as you stay within the Banana, you'll have a great time in London. Almost everything outside the Banana is horrible these days, best avoid. Not clear why, or when this happened. But it is what it is.
Taylor Lindores-Reeves retweeted
“After you have exhausted what there is in business, politics, conviviality, love, and so on—have found that none of these finally satisfy, or permanently wear—what remains? Nature remains; to bring out from their torpid recesses, the affinities of a man or woman with the open air, the trees, fields, the changes of seasons—the sun by day and the stars of heaven by night.”
— Walt Whitman
Taylor Lindores-Reeves retweeted
For those running LGTV’s, have fun
github.com/JulioFerrero/own-…
Taylor Lindores-Reeves retweeted
if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1
now you have evil qwen
recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input
We found DeepSeek V4.1 Flash 1017 beats GPT 6 Sol on our multi-step offensive cybersecurity task. Sol was tested via the Daybreak trusted access program.
We have been building a cybersecurity benchmark based on simulated corporate networks. This is a valuable indicator for the cyber-risk a model poses if in the hands of a bad actor. We plan to make the full evaluation public in the coming weeks but wanted to share our first immediate finding here.




















































