Martin Klepsch
@martinklepsch.org
Web, AI and wherever curiosity will take us :) Oh and Clojure/Script of course
I've tried a few Obsidian AI plugins but none really stuck, so I did what every sane person would do and wrote « repomix for Obsidian » Introducing `obsidian-context` github.com/martinklepsc...
GitHub - martinklepsch/obsidian-context: A CLI script to query and export content from your Obsidian vault, works fully local & read-only
A CLI script to query and export content from your Obsidian vault, works fully local & read-only - martinklepsch/obsidian-context
github.com
you gotta admire the raw-ness of @antirez.bsky.social youtube channel, lots of interesting stuff and mostly "just putting it out here" vibes 👌
I’ve fully come around to file system based routing. yes it’s weird to have some of these [slug] filenames but wow, it’s just so much less boring glueing stuff together
i think the thing about JS that bothers me the most is the laizzes-faire control flow, return anywhere, crazy syntax constructs for if expressions, … if it only was s-expressions 🥲
its becoming annoyingly commonplace that model choosers switch to 'auto' and then don't even clearly label what model was used and why 🥲 like how am I supposed to build ~intuition~
probably the easiest $12 I spent in a while. simple, plain analytics for all my sites by @drizzle.team at $12/year!
I've been using this stack for a project and it feels so much lighter than Supabase. Love Supabase but RLS and auth just try to solve for problems I don't have. - DB: Neon - Auth: better-auth - Frontend: @astro.build - Deploy: @fly.io - Jobs: Trigger
today I was at a restaurant and ordered MCP instead of MCC... that's what this has done to me
⋆ * ⋆ ✦ ✧ models <> intelligence ⋆ * ✧ context <> knowledge ✦ * ⋆ ✧ * ⋆ ⋆ *
Can't remember where I read this but thought it was an interesting take: o1-pro, gpt-4.5 and other expensive models are effectively a "drop" to see if the market can find emergent, useful behaviors. Presumably because they haven't found them internally yet.
in a world of abundant creation, focus is everything must not forget
really enjoying @ztellman.ideolalia.com's recent essays on explaining software, first the sudoku saga, now another great read
sometimes i think about just how good of a name "namespace" is its the absolute least you could put into that word
I stopped myself from vibe coding my way to some organized markdown files and instead tried Claude code. It's cool! (and slow & expensive) But mostly cool. More things should be agents.
cant wait for when claude code gets fast, really an interesting way of working with local files. it scans and loops and mostly gets it all right
at one point i questioned my desire to help people get into open source image unrelated
The custom project rules editor in Cursor is a mess, you can put this in settings.json to just use a plain markdown editor. Shoutout to @rdotpy in the Cursor Forum!
why do none of the cursor rules directories have rules files in the new format?
should I write a web UI for aider? 😆 kind of intrigued but also stop me pls
love that on Mac audio playback is interrupted when you switch between dark and light mode, dramatic
Fully back on the Perplexity train. Having access to R1, deep research, great search grounding all in one package is such a killer. Never been so fun to follow your curiosity ✨
Hey @simonwillison.net 👋 I've made a thing to better inspect the logs `llm` is creating. Especially when scripting this can make it easier to get some kind of observability :)
Whats the most lo-fi way I can go about creating prompts and evals for those? Preferably something that just runs locally, doesn't require a subscription etc.
I want to package a small node server + react app as an NPM package that you can run with `npx`. I started with `vite-express` but when running the package it won’t serve the static files… Anyone think of an example I could look at?
A super easy change I've made in how I use LLMs: editing my initial message over providing "corrections" in subsequent messages. It means less context to get confused over for the LLM and at the same time forces me to "craft a prompt" vs. "no, do it this way". Learning more!
