bckmn

Prompting

> Let's make a timeline chart of https://www.githubstatus.com/history for the past 1yr and upload it as a gist

yields [GitHub Status Incident Timeline — Feb 2025 to Feb 2026](https://htmlpreview.github.io/?https://gist.githubuserconten...)

219 total incidents across 12 full months, averaging 18.3/month. January 2026 was the worst month, and August 2025 was the calmest.

It's all really happening. Pretty much every meeting with friends touches on what their recent sightings or stories of ICE terror have been. Everyone who hasn't seen it first hand has a second hand story.

Absolutely everyone I talk to is against ICE's actions and that is the thing giving me hope that it will be defeated by the citizenry.

I think asking your questions in that form is akin to "sorting prompts" that I learned about from https://mikecaulfield.substack.com/p/is-the-llm-response-wro... and I have been using successfully when when writing code (e.g. [as a Claude code slash command](https://www.joshbeckman.org/notes/936274709)).

Essentially, you're asking the LLM to do research and categorize/evaluate that research instead of just giving you an answer. The "work" of accessing, summarizing, and valuing the research yields a more accurate result.

[Accidental Tech Podcast](https://atp.fm/) has been valuable to me for years.

Not that story specifically, but maybe a modern reinterpretation of it was [this episode of Rick and Morty](https://rickandmorty.fandom.com/wiki/Death_Crystal).

In exploring `gh` CLI extensions recently I found at least 2 devoted to being a better PR review interface. I've also seen [Trag](https://usetrag.com/) and others pop up.

Agreed. I think this is why we’re seeing a lot of new wrappers being built like Graphite.

Previously (launch): https://news.ycombinator.com/item?id=42377018

This example is based on this proposed standard (making to robots.txt or security.txt): [The /llms.txt file – llms-txt](https://llmstxt.org/)