zenlite.
Articles
Every article on zenlite, newest first. Each claim has a source and the date it was read.
The night before: a one-page summary
Field Guide, Exam prep, part 5 of 6. Your notes cut to one page for the night before, tested on free ChatGPT: it flagged the wrong line, then kept it.
5 key points, and what it leaves out
Study with AI, part 4 of 6. Key points from your own notes plus what it left out, tested on free ChatGPT: right points, a left-out list that didn't add up.
Open every source
Field Guide, Don't get fooled by AI, part 1 of 6. Free ChatGPT gave a link for every UPI fact and every fact was right, but its first link was a dead page.
Permission lists that stay small
Field Guide, Agents set up safely, part 3 of 6. A three-rule allow list in Claude Code, tested with Opus 5.5: what it let through, what it stopped, the catch.
Revision rounds: when to revisit each subject
Plan your study (medical students), part 3: revision rounds dated for four subjects, tested on free ChatGPT. Every date right; glance at a calendar once.
Weak topics from a marked mock test
Field Guide, Exam prep with AI, part 4 of 6. A marked mock to your weak topics and a plan, tested on free ChatGPT: right topics, wrong hours box.
Ask for the risky lines only
Field Guide, Code review with AI, part 3 of 6. Two risky lines in a 132-line commit, and a prompt that asks for only those. Tested in Claude Code.
Explain it like I have 10 minutes
Field Guide, Study with AI, part 3 of 6. Your own notes explained in 5 minutes of reading, tested on free ChatGPT: right, but 3 words never explained.
A week plan that survives postings and duties
Plan your study (medical students), part 2: a week plan around postings and a night duty, tested on free ChatGPT. It fits; the hours don't add up.
Make it quiz you
Exam prep with AI, part 3 of 6. Free ChatGPT quizzes you on your own notes, tested: fair marks and hints, and a revise list that missed the hinted answers.
A small diff with three problems: what the AI review catches
Field Guide, Code review with AI, part 2 of 6. Three planted bugs in a small diff, reviewed by the Claude Code CLI. Tested, with the catch.
Keep .env files and keys out of its reach
Field Guide, Agents set up safely, part 2 of 6. Three deny rules that keep Claude Code out of your .env, tested in a demo repo with fake keys, with the catch.
Flashcards from your own notes
Field Guide, Study with AI, part 2 of 6. Your own notes into ten flashcards, tested on free ChatGPT: clean cards in seconds, and the list of what didn't fit.
A month-by-month plan from your subject list
Field Guide, Plan your study (medical students), part 1. A month-by-month plan from your subjects and real hours, tested on free ChatGPT, with the catch.
Practice questions from your own notes
Field Guide, Exam prep with AI, part 2 of 6. Practice questions from your own notes, tested on free ChatGPT: the right questions, and the catch.
Read-only first: what an agent can touch
Field Guide, Agents set up safely, part 1. What Claude Code's default mode does alone and what it asks first, tested in a demo repo, with the catch.
A seeded bug: does the AI review find it?
Field Guide, Code review with AI, part 1 of 6. Plant one bug you know about, ask Claude Code to review it, and check what it finds. Tested, with the catch.
Notes to mind map in 60 seconds
Field Guide, Study with AI, part 1 of 6. Your own notes into a mind map, tested on free ChatGPT: fast and faithful, and the catch.
Your last 7 days: a revision plan from your topic list
Field Guide, Exam prep with AI, part 1 of 6. A 7-day revision plan from your own topic list, tested on free ChatGPT: what worked and the catch.
Gemini CLI 0.65.0 won't rewrite an untrusted repo's settings
From 0.65.0, Gemini CLI won't write a repo's settings file in an untrusted folder. What the fix covers, what was already true, and what to check.
A neat AI invoice can still be wrong. Five checks before you send.
Chatbots can write a tidy invoice with wrong figures. What the makers say, why a neat layout proves nothing, and five checks to run before you send.
Let the AI quiz you on your lab results. Answer first.
Ask an AI chatbot to quiz you on your own lab readings and answer before it replies. What the research says, the steps, and what to check.
Three AI habits to change at home, and what to do instead
Copying answers, trusting unchecked facts and oversharing with a chatbot can hurt your child's learning or privacy. A safer swap for each.
Ollama says 128K, NVIDIA says 64K: Nemotron 3.5 Bolt E4B context
Ollama says Nemotron 3.5 Bolt E4B has 128K of context. NVIDIA's own page says 64K. Here's what we found, and what we'd do before trusting either.
Tell any AI tool your syllabus topics before it quizzes you
Practice questions should come from your syllabus. Type your topics into any AI chatbot first and tell it to stay on that list. Three steps, any chatbot.
What can a Claude Code mod reach once you install it?
A Claude Code mod runs as you. The docs give one command that lists what it hooks, writes and sends without running it. Here is how to read it.
One shared space for text, code, images, video and audio
EmbeddingGemma 2 puts text, code, images, video and audio in one 768-number space. What that changes in search, and five steps to check it yourself.