Close the world. Open the nExt Joey (@aijoey) Lain Inspired for Hermes Agent — https://nitter.net/aijoey/status/2039688258422997126#m
PROMPT ENGINEERING
-
Chunking Strategy for Effective AI Document Processing
By
–
Use epub not PDF, convert epub to txt/md, summarize the wikipedia article into "book context", with it in context summarize one chapter at a time, etc. I mean basically imo for good results you have to "work it" in chunks and shouldn't expect that just attaching a pdf and asking
-
LLM Discord Bot Compacts Conversations into Entity Dossiers
By
–
I've been doing something similar with a Discord I'm in. The LLM reads chronologically and compacts by week, then compacts weeks to years, then years to full essays. It also keeps track of entities it encounters (people, projects, etc) and creates dossiers on them. All with
-
AI Systems Speaking MCP Protocol With Easy Training
By
–
My AI speaks MCP. And is easily trained. Just a little talking it through what we want.
-
LLMs Learn Document Patterns for Wiki Management
By
–
Atm it's not a fully autonomous process, I add every source manually, one by one and I am in the loop, especially in early stages. After a while, the LLMs "gets" the pattern and the marginal document is a lot easier, I just say "file this new doc to our wiki: (path)".
-
Hermes Agent now free with Atomic Chat
By
–
Now you can use Hermes Agent powered by Gemma 4 with Atomic Chat for free.
— 🚨 AI News | TestingCatalog (@testingcatalog) 2 avril 2026
Every local model on Atomic Chat is patched by the TurboQuant algorithm to get a larger context window and improved performance.
Hermes Agent on steroids 💪👀 https://t.co/L4brhMVwza pic.twitter.com/T3XLY12JGWNow you can use Hermes Agent powered by Gemma 4 with Atomic Chat for free. Every local model on Atomic Chat is patched by the TurboQuant algorithm to get a larger context window and improved performance. Hermes Agent on steroids
-
Developer Shares Workflow Using Replit Claude Code and Codex
By
–
yeah, i love Replit for projects like this cuz i can open up the mobile app, prompt it, click publish and share the link all from a single app I also use Claude code and codex regularly for more complex stuff
-
LLM prompt injection fails on frontier models
By
–
New report from us: Can you prompt inject your way to an “A”? As LLMs increasingly are used as judges, people are inserting AI prompts into letters, CVs & papers. We tested whether it works. It does on older & smaller models, but not on most frontier AI: https://
gail.wharton.upenn.edu/research-and-i
nsights/hidden-prompt-injections/
… -
Coding Agents Demand Deep Software Engineering Experience
By
–
"Using coding agents well is taking every inch of my 25 years of experience as a software engineer." Simon Willison (
@simonw
) is one of the most prolific independent software engineers and most trusted voices on how AI is changing the craft of building software. He co-created -
Everything Created by AI with Single Prompt
By
–
Yes and everything was created by AI with a single prompt