I always have a backup, when i travel even a 2 hour drive away from home.

I take a spare sensor, spare pump (Omnipod), spare insulin, and an insulin + dextro.

On vacation, I take a spare phone, in case my phone that manages pomp and sensor breaks down.

It happened already twice on vacation in the past 5 years.

You would need to have go installed. For my golang opensource project, also added releases on pypi and also npm.

Does this 60min ttl of also apply to claude code web?

I have regularly sessions open for multiple days.

Is that a pattern that is not advised?

Even co-pilot writes most of my code in april 2026.

Further, i don't trust code anymore that hasn't been reviewed 3x or more by co-pilot.

If you have asked me 6 months ago I wouldn't have expected this change so soon.

Writing mark down linter, i am i nterested in - if you think that you requirements for markdown formatting could be encoded in (relatively) simple rules?

Commommark is the best, then looking at the most commonly implemented extensions:

- Superscript with hat symbol (^)

- Hidden content (also known as spoiler tags, content warnings, click-to-reveal, etc...) with two pipe symbols (||) at start and end of hidden content. Interact with the content to show the content inside.

- Table syntax to show tabular data and align content with pipe, hyphen, colon symbols

---

After that, I'd look at the extensions that are only useful for the use-case you're targeting:

- Subscripting with tilde symbol (~)

- IDs, Classes and Attributes

- Task lists

- Sections / Containers

- Language labels for code blocks

- Math Support (mathml, latex, katex, etc...)

- Table of Contents / Footnotes

- Definition lists

- etc...

Just search "Markdown grammar"

Check out commonmark

Had a quick look. Stumbled upon the markdown format smd.

Was wondering if using front-matter instead of a "custom" encoding for parseble data was considered?

Yeah, I did briefly consider front-matter, but ended up with inline @ tags because I thought it kept the entire document feeling like one coherent spec instead of header-data + body, front matter felt like config to me, but this is 0.0.1 so things might change :)

Interesting - what kind of structural issues have you encountered?

Is these more related to the existing source code or is this a bad pattern thar you would never do regardless of the existing code?

In Chess this has been going on for a while. Story of humans playing Chess is still entertaining - while AI making amazing moves seems to be less news worthy in my perception.

There are frameworks like https://github.com/bmad-code-org/BMAD-METHOD and https://github.github.com/spec-kit/ that are working on encoding a similar kind of approach and process.

Claudes search tool uses Ripgrep , which is embedded in Claude.